INTRODUCTION Electroconvulsive therapy (ECT) uses a small electric current to produce a generalized cerebral seizure under general anesthesia.

. ECT is used mainly to treat severe depression, but is also indicated for patients with other conditions, including bipolar disorder, schizophrenia, schizoaffective disorder, delirium, and neuroleptic malignant syndrome. There is no question about the efficacy and safety of ECT, which is practiced widely in the United States and the rest of the world . Nevertheless, it remains controversial and stigmatized because of misinformation and outmoded perceptions about how the treatment is performed. This topic provides an overview of ECT. The indications for treating unipolar major depression with ECT, the efficacy of ECT for treating unipolar depression, technique for performing ECT, and medical consultation for ECT are discussed separately. HISTORY OF ECT The use of seizures to treat psychiatric disorders has progressed over time, and began with observations that schizophrenic patients often improved temporarily after a spontaneous seizure. Initially, physicians induced seizures with medications. In 1938, Italian psychiatrists used an electric current to induce seizures as a treatment for schizophrenia. ECT quickly spread around Europe and the rest of the world, and was first used in the United States in 1939. Patients suffered bone fractures occasionally and considerable anticipatory anxiety, but advances in anesthesia techniques in the 1950s allowed for the use of general anesthesia and muscle relaxation for ECT, which reduced the incidence of these problems. The use of ECT in the United States has fluctuated. Use of ECT declined in the 1970s, and subsequently increased, possibly as a result of greater treatment resistance to pharmacotherapy in depressed patients, increased recognition of the limitations of pharmacotherapy, and better public acceptance. A practice survey from 1988-89 estimated that at least 100,000 patients received ECT annually. The typical ECT patient today is relatively affluent and receives ECT in a private sector psychiatric facility. State hospitals rarely offer the treatment, even though many of the patients would meet indications for ECT. MECHANISM OF ACTION The mechanism of action for ECT is unknown, but there are a myriad of well-documented changes in the central nervous system: Human and animal studies show that ECT increases release of monoamine neurotransmitters, particularly dopamine and serotonin. ECT also enhances monoamine transmission by desensitizing presynaptic adrenergic autoreceptors. One theory postulates that ECT exerts its beneficial effects by increasing release of central neuropeptides, including corticotrophin releasing factor, somatostatin (STS), and neuropeptide Y. The neuroendocrine hypothesis is based upon the diabetes/insulin model and suggests that mood disorders are caused by insufficiency of a hypothalamic mood-maintaining peptide. Repeated seizures enhance the production and release of the putative hypothalamic peptide

antidepression, which relieves both neuroendocrine and behavioral abnormalities. ECT has anticonvulsant properties, which has led to the suggestion that these properties are responsible for the therapeutic effects of the treatment. Positron emission tomography (PET) studies demonstrate decreased metabolic activity in frontal and cingulate cortex after ECT. Functional magnetic resonance imaging before and after successful treatment with ECT reveals a reduction in global connectivity within the left dorsal lateral prefrontal cortex. Quantitative electroencephalogram (EEG) studies demonstrate increased slow (delta) wave activity in the prefrontal cortex after ECT, which is associated with clinical response. Several human and animal studies indicate that ECT has trophic effects on the central nervous system. ECT increases brain derived neurotrophic factor (BDNF) in patients and electroconvulsive stimulation induces neurogenesis and mossy fibre sprouting from granule cells in rat hippocampus. PRE-ECT EVALUATION The goals of the pre-ECT evaluation are to: Determine whether ECT is indicated. Establish baseline psychiatric and cognitive status to serve as reference points for evaluating response and cognitive effects. Identify and treat any nonpsychiatric medical conditions that might be associated with an increased risk of adverse events from ECT. Initiate and continue the informed consent process. A complete medical history should be taken. Medical consultation prior to ECT is indicated for most patients 40 years or older, and is frequently obtained in younger patients as well. Particular attention should be paid to any history of cardiopulmonary disease and prior surgeries (with inquiry about type of anesthesia and any complications). A history of skull fractures should be determined, as this may affect electrode placement. Appropriate physical examination and laboratory evaluation, guided by the relevant history, should be performed. The patient's handedness and any dental problems should be noted. Specific laboratory tests are not indicated for patients without known medical comorbidities. SPECIFIC MEDICAL CONDITIONS Inducing a seizure causes transient increases in blood pressure, pulse, and intracranial pressure, which can have deleterious effects. Organ systems of most concern are: Cardiovascular Pulmonary Central nervous system Seizures increase cardiac workload and oxygen demand. Thus, it is important to try to

maximize the patient's cardiovascular status before ECT. Additionally, medication to blunt increases in heart rate and blood pressure is indicated prior to ECT and during the procedure in patients for whom these increases would be detrimental. Coronary heart disease Cardiac complications are the most common cause of serious morbidity and mortality during ECT. Patients with coronary heart disease should be evaluated prior to ECT with a careful cardiac history and ECG. Additional cardiovascular diagnostic testing and evaluation by a cardiology consultant may be indicated for selected patients, and for any patient with unstable disease. History of myocardial infarction History of myocardial infarction and current unstable angina present increased risk. Careful consideration of the risk/benefit ratio of ECT in such cases is essential. The key factors in this assessment are the extent of myocardial damage and subsequent healing, and the residual functional cardiac status. In general, the longer one waits after a myocardial infarction, the safer it is to perform ECT. For patients who have had a myocardial infarction and are not at imminent risk from their depressive episode, we suggest waiting a minimum of three months prior to starting a course of ECT. However, patients at high risk of death due to their depression can be treated sooner. For severely depressed patients, treatment with ECT on an emergent basis may be indicated. Hypertension The patient's blood pressure should be well controlled prior to ECT. The patient should take his or her currently prescribed antihypertensive medications (with the exception of diuretics) with a small sip of water, prior to each ECT treatment. Use of prophylactic beta blockers to prevent treatment-induced hypertension is discussed separately. Heart failure and valvular disease The use of ECT in patients with heart failure or valvular disease is discussed separately. Pacemakers and implantable defibrillators The use of ECT in patients with pacemakers or implantable defibrillators is discussed separately. Pulmonary disease Optimizing pulmonary status before ECT is important. The anesthesiologist manipulates the patient's airway during ECT using a bag and mask circuit and there is a risk of inducing bronchoconstriction. All patients should receive supplemental oxygen via nasal cannula, initiated a few minutes prior to each treatment.

 Elective intubation is not indicated for ECT, except in rare circumstances in which the anesthesiologist assesses that intubation would be exceptionally difficult should the airway become compromised. Neurologic disease The use of ECT for patients with brain tumors, history of stroke, dementia, and neuromuscular disease is discussed separately CONCURRENT MEDICATIONS The patient's current medications, including prescription, over-the-counter, and complementary or alternative drugs should be reviewed prior to ECT. The ECT psychiatrist, anesthesiologist, or medical consultant should decide which to continue and which to taper and discontinue. Psychotropic drugs Many psychotropic medications may be continued during a course of ECT for their synergistic effect without compromising safety, including antidepressants, antipsychotics, and lithium. Morning doses on the day of ECT are given after the patient has recovered from that day's procedure. Anticonvulsants and benzodiazepines often interfere with ECT and may need to be tapered and discontinued. Antidepressants improve the efficacy of ECT. A review of 10 systematic reviews, 7 metaanalyses, and 3 practice guidelines found that for depressed patients, a tricyclic may improve the antidepressant effect of ECT. As an example, a randomized trial assigned 319 depressed patients receiving an acute course of ECT to concomitant nortriptyline, venlafaxine, or placebo. Remission occurred in more patients who received nortriptyline or venlafaxine compared with placebo (63 and 60 versus 49 percent) with a significant difference between nortriptyline and placebo. Antidepressants are generally safe to use with ECT and do not affect its tolerability. As an example, a randomized trial of depressed patients receiving an acute course of ECT with concomitant nortriptyline, venlafaxine, or placebo found the number of adverse events did not differ significantly between the three groups . Other types of antidepressants are also well tolerated during ECT, including selective serotonin reuptake inhibitors, tetracyclics, and monoamine oxidase inhibitors. Antipsychotic medications are often continued during ECT and may provide a synergistic antipsychotic effect, especially second generation antipsychotics. In addition, antipsychotics generally do not appear to affect the tolerability of ECT. Second generation antipsychotics and high potency first generation antipsychotics are thought to be safe. A review found that early reports warned of hypotension and cardiac complications from combining chlorpromazine and ECT, but later reports did not substantiate these problems . Nonetheless, combining of low potency, first generation antipsychotics with ECT should be done cautiously.

Lithium has the potential to increase the adverse cognitive effects of ECT and to prolong the effects of succinylcholine (which is typically used during ECT to reduce tonic-clonic movements), but these problems are rare . However, a review of several case series found that concomitant lithium and ECT is safe [33]. We suggest reducing lithium levels below the full therapeutic range at the time of each ECT treatment by withholding one or two doses of lithium prior to each treatment. Benzodiazepines should be tapered and discontinued whenever possible during ECT because of their anticonvulsant properties, which can increase seizure threshold, shorten seizure duration, and decrease the intensity of the ECT seizure [34]. These effects may decrease the clinical efficacy of ECT. For patients who need to continue their benzodiazepine during ECT, some authorities recommend switching from drugs with a long half-life to one with a short half-life, and withholding the evening dose before each ECT treatment. Small doses of benzodiazepines, however, may not be a problem. Patients who are very anxious about the ECT procedure itself may take a dose of benzodiazepine (eg, 1 mg of lorazepam sublingual) 30 to 60 minutes before the treatment. Anticonvulsants may interfere with the efficacy of ECT, similar to benzodiazepines [35]. When anticonvulsants are used for mood stabilization, they generally should be tapered and discontinued prior to ECT. Patients with epilepsy should continue their anticonvulsants during a course of ECT. Clinicians should consider reducing the dose if seizures are difficult to elicit, or withholding a dose the evening or night before the ECT treatment. Other drugs Cardiovascular medications Patients should receive their routine antihypertensive (other than diuretics) and antianginal medications with a small sip of water approximately two hours before ECT. Diuretics should not be given because it is better to minimize the amount of urine in the bladder during the procedure, to prevent the patient from wetting himself as a result of the seizure. Asthma medications Glucocorticoids and beta-adrenergic agonists may be given before ECT, as needed, to prevent bronchoconstriction. Patients who routinely use a fast-acting inhaler should do so immediately before the treatment. Theophylline should be avoided or maintained at the lowest effective blood level because it has been associated with prolonged seizures and status epilepticus [36]. Gastrointestinal medications Patients with gastroesophageal reflux disease (GERD) should

receive their H2 blocker, proton pump inhibitor, or metoclopramide with a sip of water two hours before ECT to prevent reflux and possible aspiration. Sodium citrate (30 cc po) may be given immediately before treatment to neutralize any acid remaining in the stomach [29]. INFORMED CONSENT The psychiatrist should ensure that the consent process conforms to all state and local laws and statutes, and hospital policies. Informed consent requires that the patient [29]: Receive adequate information about depression and ECT Is capable of understanding and acting reasonably on this information Is given the opportunity to consent in the absence of coercion ECT consent forms are more detailed compared with those for most other medical procedures because ECT is closely scrutinized.

Informational material written for the layperson is typically provided to patients and family members as part of obtaining informed consent. In addition, patients and family members may watch video to augment the written material and discussion with the psychiatrist recommending ECT. The vast majority of ECT is done with fully informed consent given by the patient. In circumstances when the patient is too ill (eg, catatonic or psychotic) and lacks capacity to provide consent, and the need for ECT is urgent or emergent, the psychiatrist should seek substitute consent by a court. The patient should be cautioned not to make major personal or financial decisions during or immediately after a course of ECT. In addition, the patient should not drive until the cognitive effects of ECT have resolved. Patients who are receiving continuation or maintenance ECT should not drive until 24 hours after each treatment [37]. TREATMENT COURSE Number of treatments There is no standard number of treatments for an acute ECT course and no way to predict how many treatments a particular patient will need. Most patients remit with 6 to 12 treatments, but some require only three while others require 20 or more [29]. In a study of 184 patients with unipolar major depression, the mean number of treatments needed to reach remission was 7 (standard deviation 3) [38]. A second study of 230 patients with unipolar or bipolar major depression found that 90 percent or more of the remissions occurred with nine or fewer seizures, depending upon the type of electrode placement [39]. Acute ECT should last until the patient remits, reaches an improvement plateau, or develops limiting adverse effects. It is becoming more common to taper the frequency of ECT treatments,

rather than abruptly stop a course. Once remission is achieved, however, ongoing continuation and maintenance treatments are frequently indicated. Treatment frequency The frequency of treatments varies by country and the clinical urgency. Standard practice in the United States is to give ECT three times per week on a MondayWednesday-Friday schedule. The routine in many other countries is twice a week, particularly for elderly patients. Urgently ill patients (eg, catatonic or severely malnourished) may be given daily bilateral ECT until some improvement is evident [37]. Meta-analyses of five randomized trials (169 patients with unipolar major depression) found that symptomatic improvement, remission, and number of ECT treatments were comparable for twice-weekly and thrice-weekly ECT; however, duration of treatment was significantly longer for twice-weekly ECT (by nearly five days) [40]. The data were insufficient to analyze differences in cognitive impairment. Clinical monitoring Depressive symptoms and cognition should both be monitored. Several depression rating scales are available; the most common is the Hamilton Rating Scale for Depression (HRSD or "HAM-D") [41]. The scale should be administered at baseline, weekly during the course of ECT, and when the ECT course is completed. Many scales are also available for monitoring cognition; the most common is the Mini Mental State Examination [42]. It is easy to use and adequate for routine clinical use, although its sensitivity may be limited [43]. More comprehensive neuropsychologic testing may be indicated for selected patients (eg, those with known cognitive deficits at baseline). ADVERSE EFFECTS The adverse effects of ECT may be divided into medical or cognitive effects. Adverse medical effects The mortality rate of ECT is 2 to 4 deaths per 100,000 treatments, making it one of the safest procedures performed under general anesthesia [44,45]. A review of 10 systematic reviews, 7 meta-analyses, and 3 practice guidelines found that the mortality associated with ECT was comparable to the mortality associated with minor procedures involving general anesthesia [27]. Concerns that ECT causes structural brain damage have been dispelled by multiple human and nonhuman primate studies [1,27]. Mortality is mostly related to cardiopulmonary events. Patients with coronary heart disease may be at risk for cardiac ischemic events because the seizure increases cardiac workload and oxygen demand. The cardiovascular effects of ECT are discussed separately. Other adverse medical effects include:

 Aspiration pneumonia Risk is increased in patients who do not have an empty stomach . Fracture Patients with severe osteoporosis are at increased risk of fracture, and extra care should be taken to insure excellent muscle relaxation [46]. Dental and tongue injuries These may occur when the oral bite block is not fully protective [47]. Loose teeth are at risk for dislodgement during the procedure and possible aspiration, and should be stabilized or extracted before ECT. Headache This is the most common, nonserious adverse medical effect of ECT [48-50]. Patients should be told to expect some headache after each treatment. It is typically managed with acetaminophen or ibuprofen given after the treatment. Prophylaxis with IV ketorolac 30 mg may be considered for patients with significant post-ECT headache [51]. Nausea Transient, postprocedure nausea is common and is the result of the anesthesia and airway manipulation, which may introduce air into the stomach. Prophylaxis with IV ondansetron 4 mg should be given to patients with significant post-ECT nausea. Other somatic symptoms Patients may complain of other somatic symptoms such as myalgias. However, one study assessed a group of depressed patients pre-ECT and during an acute course of ECT, and found the frequency of somatic symptoms other than headache did not increase [52]. The authors suggested that these other somatic symptoms may have been due to the depressive illness, the use of psychotropic medications, or withdrawal from such medications. Adverse cognitive effects Most patients report some adverse cognitive effects during and after a course of ECT. The incidence depends upon electrode placement, stimulus type and dose, anesthesia, and the patient's pretreatment cognitive status. A systematic review found four studies in which the proportion of patients who reported any memory loss ranged from 51 to 79 percent [53]. ECT causes three types of cognitive impairment: Acute confusion Anterograde amnesia Retrograde amnesia The acute confusional state is the result of both the seizure and the anesthesia. It typically resolves 10 to 30 minutes after the procedure. Anterograde amnesia is the decreased ability to retain newly acquired information. It occurs during a course of ECT and typically resolves within two weeks after completing the course. We thus suggest that patients refrain from driving for 24 hours after an ECT treatment or from making important business or personal decisions one to two weeks after completing an acute

course of ECT [54]. Retrograde amnesia involves forgetting recent memories and is the most anxiety-producing cognitive effect of ECT. The memories are for events that occur during the course of ECT and a period of weeks to a few months prior to that. The deficits are greatest and most persistent for knowledge about public or world events (impersonal memory) compared with knowledge about the self (personal memory) [55]. Bilateral ECT causes more retrograde amnesia than right unilateral. Retrograde amnesia recovers more slowly than anterograde amnesia [55-57]. Some of the lost memories of events prior to the course of ECT may be expected to return, while others may not. A systematic review found four studies in which the proportion of patients who reported persistent or permanent memory loss ranged from 29 to 55 percent [53]. The etiology of these more severe memory impairments is controversial [58,59]. Regardless, the typical ECT patient is profoundly depressed and accepts some degree of memory loss as a reasonable tradeoff for resolution of the depressive episode. Impaired cognition due to depression often improves after a course of ECT [60]. A review of ECT in elderly patients with impaired cognition due to depression ("pseudodementia") examined six studies that tested cognition before and after ECT in patients mostly 55 years or older [61]. Patients were assessed with a cognitive screening instrument, either the Mini Mental State Examination [42] or the Mattis Dementia Rating Scale [62]. Global cognitive functioning in patients with cognitive impairment improved in all studies. CONTINUATION AND MAINTENANCE ECT Continuation ECT is the practice of providing a single ECT treatment, at an interval of one to eight weeks, during the first six months after remission. The purpose is to prevent relapse of the mood or psychotic episode that prompted the acute course of ECT. Maintenance ECT refers to ECT given beyond continuation ECT, to patients who have recovered from the acute mood or psychotic episode but require treatment to prevent recurrence of a new episode. Continuation and maintenance ECT are usually outpatient procedures. Mood and psychotic disorders are chronically recurring illnesses. Therefore, continuation and maintenance treatment (with pharmacotherapy, ECT, psychotherapy, or a combination) are required for most patients who recover from an initial episode. As an example, in a prospective, observational study of 154 patients with major depression who remitted after successful treatment with ECT in community hospitals, 64 percent relapsed during 24 weeks of follow-up [63]. We suggest continuation and maintenance ECT for depressed patients who have successfully completed a course of acute ECT and: Relapse without continuation and maintenance treatment (either pharmacotherapy or ECT), and

require a second a course of acute ECT Fail one or more courses of continuation and maintenance pharmacotherapy Prefer continuation and maintenance ECT The electrode placement used for continuation and maintenance ECT should be the same as that used for acute ECT. Pharmacotherapy is often combined with continuation and maintenance ECT, with improved outcomes [64]. The schedule for continuation and maintenance ECT generally starts with weekly treatments for several weeks, and gradually tapers to the maximal interval that allows the patient to remain euthymic. Following an acute course of ECT three times per week, we suggest treatments be administered [38]: Weekly for four weeks Biweekly for eight weeks Monthly for two months Some patients continue to taper down to one treatment every three months, depending upon their clinical status and preference. Patients who benefit from maintenance ECT can receive it indefinitely. Use of self report and clinician administered rating scales to monitor the patients clinical status is discussed separately. Following acute ECT, continuation ECT alone appears to be comparable to continuation pharmacotherapy alone. A randomized trial compared continuation treatment with either ECT or pharmacotherapy in 184 patients who remitted from unipolar major depression after an acute course of bilateral ECT [38]. Patients were assigned to a fixed schedule of continuation ECT or combination pharmacotherapy with lithium plus venlafaxine. In both groups, 46 percent of patients remained euthymic for the six months of the study. In addition, the two continuation treatments did not differ in memory effects [65]. Cognitive effects of continuation and maintenance ECT are generally minor or nonexistent: A controlled trial randomly assigned depressed patients who remitted with an acute course of ECT to continuation ECT or continuation pharmacotherapy, and evaluated them with an extensive battery of memory tests at baseline and after 24 weeks of treatment [65]. There were no significant differences between groups on any of the memory tests. An observational study evaluated 20 patients (multiple diagnoses) receiving maintenance ECT with a comprehensive neuropsychologic battery and retested them one year later [66]. There were no significant differences in test scores over time. In addition, test scores for the patients receiving maintenance ECT did not differ significantly from a control group of patients not receiving ECT.

STIGMA, ACCESS, AND TRAINING Stigma Stigma remains the biggest impediment to the appropriate use of ECT. The level of knowledge about modern ECT among patients and medical practitioners is low [67]. The portrayal of ECT in the media as a crude and coercive treatment remains problematic, frightening off patients who might benefit from it. Improved education about ECT among health care professionals will lead to more favorable perceptions about its use [68]. Access Access to ECT in the United States is variable and very limited in some rural areas. It is rarely available to patients in state psychiatric hospitals. Patients with no or limited insurance coverage may be unable to pay for ECT, leading to a situation in which ECT is preferentially available to more affluent patients with better insurance [69]. Training ECT is a standard part of training in general psychiatry residency programs in the United States, and many programs offer in-depth hands-on training. Guidelines for curriculum are suggested by the American Psychiatric Association [29]. Several post-residency hands-on training courses are offered throughout the United States. Individual United States hospitals issue privileges to psychiatrists to perform ECT, but there is no board certification for ECT in the United States [1].

A typical course of ECT consists of 6 to 12 treatments, individualized for each patient. Treatment should usually continue until remission of mood or psychotic symptoms Standard practice in the United States is to give ECT three times per week. The routine in many other countries is twice a week. Continuation ECT is the practice of providing a single ECT treatment, at an interval of one to eight weeks, during the first six months after remission. Maintenance ECT refers to ECT given beyond continuation ECT. Continuation and maintenance ECT are usually provided on an outpatient basis to prevent relapse or recurrence of the mood or psychotic episode that prompted the acute course of ECT. The mortality rate of ECT is 2 to 4 deaths per 100,000 treatments, making it one of the safest procedures performed under general anesthesia. Mortality is mostly related to cardiopulmonary events. Other adverse medical effects include aspiration pneumonia, fracture, dental and tongue injuries, headache, and nausea. Most patients report some adverse cognitive effects during and after a course of ECT, including acute confusion (delirium), anterograde amnesia, and retrograde amnesia. SETTING ECT is performed in a dedicated ECT treatment suite, a hospital post-anesthesia care unit, or an ambulatory surgery suite.

ECT may be performed on an inpatient or outpatient basis, and there is an increasing trend toward outpatient ECT. Patients with severe medical or psychiatric illness may start ECT on an inpatient basis and as they improve, switch to outpatient treatment. Continuation and maintenance ECT are almost always given on an outpatient basis. PRE-ECT ORDERS The orders for each ECT treatment include: No food or drink (NPO) after midnight Intravenous line The patient receives general anesthesia as part of ECT and thus should not eat solid food for six to eight hours prior to the procedure, and not drink clear liquids for two hours, except for necessary medications with a small sip of water [2,3]. The patient should empty his or her bladder before the treatment. TREATMENT PROTOCOL ECT is administered by a team that typically includes a psychiatrist, anesthesiologist, and nurse. Most patients are treated according to a protocol that addresses the following issues [3]: Patient safety Confirm the identity of the patient, that the patient is NPO, has taken necessary medications, and signed the informed consent form. In addition, the patient is monitored after the seizure for approximately 30 to 60 minutes. Equipment ECT device, device to monitor vital signs and pulse oximetry, anesthesia equipment, nasal cannula to provide oxygen, nerve stimulator to assess degree of motor relaxation, electromyograph, blood pressure cuffs (arm and ankle) Medications Anesthetic, neuromuscular blocker. Anticholinergic and cardiovascular medication as indicated Documentation Record vital signs, ECT device settings, and duration of seizure PHYSIOLOGIC MONITORING Physiologic functions monitored during ECT include: Vital signs Blood oxygen saturation Cardiac rhythms (ECG) Electrical activity of the brain (EEG) In addition, electromyography (EMG) may be performed to evaluate the electrical activity of the muscles and a nerve stimulator may be used to monitor the effects of succinylcholine (used to reduce tonic-clonic contractions). EEG monitoring is crucial because it enables the practitioner to confirm that a cerebral seizure has occurred and has ended in a timely fashion. The EEG is typically recorded with one or two

channels, from right and left frontal and mastoid positions (figure 1). ECT devices display the EEG tracing on a paper record and/or computer screen, and may also provide information about the adequacy of the recorded seizure by means of automated indices. Despite the device interpretation, interpretation of the EEG remains an important skill for the ECT practitioner. EMG is generally recorded from the right foot in order to measure the length of the motor component of the seizure. This corresponds with the possible use of right unilateral electrode placement. A blood pressure cuff is placed and inflated on the right ankle to prevent succinylcholine from entering the foot. An alternative to using the EMG is to simply place a blood pressure cuff on the ankle and have a member of the treatment team watch the foot after the electric stimulus is administered and record the duration of tonic-clonic contractions. ELECTRODE PLACEMENT The electric current used to induce the seizure passes briefly through the brain via two electrodes applied to the scalp (using either the adhesive type of electrode or hand-held type). Psychiatrists typically position the stimulating electrodes in one of three ways (figure 2): Bilateral Right unilateral Bifrontal Bilateral, also known as bitemporal, is the original, "gold standard" placement. One electrode is placed on each temple, with the center of the electrode placed 2 to 3 cm above the midpoint of the line connecting the outer canthus of the eye and the external auditory meatus on each side of the head. Bilateral placement has the greatest antidepressant efficacy and quickest speed of response, but may cause the most memory impairment. Right unilateral electrode placement consists of placing one electrode on the right temple and one on the scalp, just to the right of the vertex (d'Elia placement). Specifically, one electrode is positioned as in bilateral on the right side, and the center of the other electrode is placed 2 to 3 cm to the right side of the vertex of the skull. This technique avoids initial stimulation of the left cerebral hemisphere, which is usually dominant for language functions. Right unilateral may be slightly less effective for some patients, but generally causes fewer adverse cognitive effects such as memory impairment. Bifrontal consists of placing each electrode on the forehead above the outer canthus of each eye. The center of each stimulating electrode is placed 4 to 5 cm above the outer canthus of the eye along a vertical line perpendicular to a line connecting the pupils. Some clinicians believe that bifrontal is as effective as bilateral with the same cognitive impairment as right unilateral, although there is a much smaller evidence base for this technique. Nonetheless, it is used widely in clinical practice. A meta-analysis of 22 randomized trials (1408 patients) found that although bilateral ECT was

superior to unilateral ECT for treating depression, the difference was small and likely to be of only minimal to moderate clinical significance [4]. A subsequent randomized trial compared the three electrode placements in 230 patients with unipolar or bipolar major depression and found no statistically significant difference in the proportion of patients who remitted with bilateral, bifrontal, or right unilateral placement (64 versus 60 versus 55 percent) [5,6]. Bilateral ECT caused a significantly better response early in the course of treatment, compared with bifrontal and right unilateral. In addition, there was little difference in the adverse cognitive effects of each placement, a somewhat surprising finding that may be attributable to the stimulus dosing of right unilateral at six times seizure threshold. Thus, we suggest: Bilateral ECT in more urgent clinical situations Right unilateral ECT when there is a significant concern about minimizing retrograde amnesia Patients whose symptoms do not respond to right unilateral ECT can be switched to bilateral. We suggest doing so after three to five treatments, depending upon the patient's clinical status and preference. Clinicians should avoid stimulating over or adjacent to a skull defect, regardless of the type of electrode placement [2]. Thus, the pre-ECT evaluation should address whether there is a history of skull fracture, which may indicate the need for skull radiographs. STIMULUS For many years it was assumed that the electric stimulus was relevant only for its ability to elicit a grand mal seizure. It is now clear that the type and intensity of the stimulus contribute to the efficacy and adverse cognitive effects of ECT [7,8]. Stimulus type The type of stimulus used in contemporary ECT devices is a brief pulse (0.5 to 2.0 milliseconds) or ultra-brief pulse (<0.5 milliseconds) waveform. Brief pulse is considered standard, but there is accumulating evidence that ultra-brief pulse stimuli may cause less cognitive impairment [9]. The outmoded sine wave form of stimulus causes more cognitive impairment and should no longer be used [8]. Stimulus dose The intensity or dose of the electric stimulus affects efficacy, speed of response, and adverse cognitive effects. Stimulus intensity is usually described in terms of seizure threshold, which may be estimated or determined empirically at the first treatment session [10,11]. Establishing the seizure threshold is accomplished by initially setting the ECT device to use a small amount of energy to induce the seizure. If this dose does not produce a seizure, the device is then set at an incrementally higher dose in an attempt to induce a seizure. This process is repeated until a seizure occurs, which then establishes the seizure threshold. The stimulus dose depends upon the electrode placement. The suggested stimulus dose for bilateral or bifrontal ECT is 1.5 to 2 times seizure threshold. The suggested dose for right

unilateral is 6 times seizure threshold, based upon data showing that seizures need to be elicited with stimuli that are several times greater than seizure threshold for unilateral placement to be maximally effective [7,12]. The higher dose required for right unilateral occasionally becomes a problem because ECT devices manufactured in the United States are limited to a charge output of 600 millicoulombs by Food and Drug Administration regulations. This prevents a small number of patients with a relatively high seizure threshold from being treated at fully six times seizure threshold [13]. It is common for seizure threshold to increase during an acute course of ECT. This may necessitate incrementally increasing the dose of the electric stimulus periodically. Clinicians should follow the specific instructions of the ECT device as specified by the manufacturer. ANESTHESIA TECHNIQUE The goal of anesthesia for ECT is to provide the patient with a safe, comfortable experience. The anesthesiologist preoxygenates the patient and administers medications to induce unconsciousness and to relax or paralyze skeletal muscles. In addition, an anticholinergic drug may be given prior to ECT to prevent bradycardia or asystole. Preoxygenation The patient is preoxygenated with supplemental oxygen (2 L/min) via nasal cannula while the procedure is being set up, with the goal of maintaining oxygen saturation at or near 100 percent. This makes ECT safer with fewer adverse effects than either a spontaneous (epileptic) seizure or ECT in the past. In addition, the patient is often hyperventilated immediately prior to delivering the electrical stimulus. This induces cerebral hypocarbia, which increases seizure intensity [14]. Anticholinergic medication Premedication with an anticholinergic agent, either glycopyrrolate (0.2 mg, IV) or atropine (0.4 mg, IV), is often used to prevent vagally-mediated bradycardia and excess oral and respiratory secretions. Bradycardia is most common at the first treatment, when there is a greater chance of administering a subconvulsive stimulus that leads to vagal outflow unopposed by subsequent sympathetic activity from a seizure. Anticholinergics may cause or exacerbate tachycardia. Anesthetic medication The anesthesia should not interfere with inducing an effective grand mal seizure. The induction agent of choice is methohexital, generally given at a dose of 0.75 to 1 mg/kg. Propofol is a commonly used alternative, but it is more potently anticonvulsant than methohexital and may reduce seizure duration in some circumstances. Other induction agents include thiopental, etomidate, and ketamine [15]. There has been interest in the use of ultrashort-acting narcotics such as remifentanil to replace at least some of the standard induction agent, but this has not been widely adopted [16].

Muscle relaxation medication Skeletal muscle relaxation is used during ECT to minimize the motor seizure and prevent musculoskeletal injury. This is particularly important for patients with osteoporosis. The standard agent for muscle relaxation is succinylcholine, a depolarizing neuromuscular blocker, given at a dose of 0.75 to 1 mg/kg via intravenous infusion. Muscle relaxation subsides promptly with discontinuing the infusion. Nondepolarizing muscle relaxants such as atracurium, mivacurium, or rocuronium are used only in special circumstances when succinylcholine cannot be used. Such circumstances include pseudocholinesterase deficiency, severe neuromuscular disease or injury (quadriplegia, amyotrophic lateral sclerosis, or muscular dystrophy), severe muscular rigidity, or recent severe and widespread burns [2,15]. Pseudocholinesterase is needed to metabolize succinylcholine, and its absence will lead to prolonged apnea and muscle relaxation. A pseudocholinesterase level (also referred to as dibucaine number) is measured only in patients with a personal or family history of prolonged apnea following exposure to muscle relaxants. It is important to ascertain that the patient is unconscious before proceeding with the muscle relaxant, to avoid unnecessary patient distress. The patient is rendered apneic by the muscle relaxant, at which point the anesthesiologist provides ventilation by bag and mask with 100 percent oxygen. The sufficiency of muscle relaxation is determined using a nerve stimulator and observing the decrement and eventual disappearance of response. In addition, the clinician can assess knee and plantar withdrawal reflexes, and can simultaneously observe for fasciculations in the calves and left foot and wait until they subside, usually within one to two minutes of administering the muscle relaxant. Perfusion of the right foot with the muscle relaxant is prevented by an inflated blood pressure cuff on the right ankle. Muscle contractions often occur despite the use of a muscle relaxant. As an example, masseter muscles may contract, and thus it is important to remove dentures, if present, prior to administering the electric stimulus. In addition, a member of the treatment team should insert a bite block to protect the teeth, making sure that the tongue is pushed inferiorly and posteriorly in the mouth, and that the chin is held firmly against the bite block. Cardiovascular medication Prophylactic cardiac medications immediately before or during ECT may be administered for the purpose of blunting the hypertensive, tachycardic response to the seizure and reducing the risk of myocardial ischemia. The most commonly used drugs in this situation are beta blockers. The use of beta blockers for this purpose is discussed separately. SEIZURE DURATION The minimal duration for a therapeutic ECT seizure is 15 seconds. Beyond this, the efficacy of ECT is not related to seizure duration [2].

Most therapeutic ECT seizures last 15 to 70 seconds on EEG recording. The recording typically lasts 10 to 30 percent longer than the motor seizure [17]. Problems with ECT include [2]: Missed seizures the electrical stimulus does not induce a seizure Short seizures last less than 15 seconds Prolonged seizures last longer than two to three minutes Short seizures may be subtherapeutic, and prolonged seizures may be associated with increased cognitive impairment. Missed seizures should be followed by a brief (eg, 20 seconds) period of hyperventilation and restimulation at an incrementally higher stimulus dose. Short seizures should be followed by a longer period of hyperventilation (eg, 60 seconds) and restimulation at an incrementally higher stimulus dose. A prolonged seizure is a potentially serious complication that needs to be recognized and treated promptly. We suggest terminating seizures that last longer than 180 seconds. The standard procedure is to administer one-half the induction dose of the anesthetic (when it is either methohexital or propofol). Alternatively, a dose of benzodiazepine (eg, diazepam 5 mg IV) may be given [2]. Managing missed or short seizures There are several procedures to use for persistent missed or short seizures [2,3,18,19]: Decreasing or discontinuing anticonvulsant mood stabilizers and benzodiazepines, if possible Hyperventilating the patient before and during the seizure Decreasing the anesthetic dose to the minimum compatible with full unconsciousness Switching anesthetic to etomidate (0.15 to 0.30 mg/kg IV) or ketamine (1 to 2 mg/kg IV), which are less anticonvulsant than methohexital Intravenous caffeine has been used to prolong seizures, but is no longer recommended because its clinical benefits are uncertain. PREGNANCY Many aspects of administering ECT to pregnant patients are comparable to the technique used for patients who are not pregnant. As an example, the electrical stimulus dose is not adjusted for pregnancy because it is not clear that pregnancy changes seizure threshold during ECT [20,21]. Electrical stimulus dosing is discussed elsewhere in this topic. (See 'Stimulus' above.) However, ECT technique for pregnant patients should be modified as follows [21-23]:

Avoid hyperventilating the patient (hyperventilation can diminish fetal oxygenation by decreasing placental blood flow and reducing the dissociation of oxygen from hemoglobin) Hydrate patients with intravenous fluids prior to each ECT treatment to minimize the risk of premature contractions For pregnancies between 12 and 23 weeks of gestation, document the fetal heart rate before and after each ECT treatment When the gestational age exceeds 20 weeks, place a wedge beneath the patients right hip during each ECT treatment to displace the uterus from the aorta and vena cava and thus optimize maternal venous return, cardiac output, blood pressure, and uterine blood flow. For pregnancies 24 weeks of gestation (the fetus is considered viable if born at this gestation age), continuously monitor the fetal heart rate and uterine activity for 30 to 60 minutes before and after each ECT treatment. A clinician with experience placing monitoring devices and interpreting these tracings should be available. Administer ECT at facilities with resources for treating obstetric and neonatal emergencies Pregnant patients are at increased risk for gastric reflux and aspiration pneumonitis, which can be minimized with one or more of the following measures during each ECT treatment [2022,24]: Withhold anticholinergic drugs Glycopyrrolate or atropine may decrease lower esophageal sphincter tone and thus increase reflux. However, for pregnant patients who require an anticholinergic drug to prevent excessive bradycardia, we suggest glycopyrrolate, which does not cross the placenta as readily as atropine. Use of anticholinergic medication during ECT is discussed elsewhere in this topic. (See 'Anticholinergic medication' above.) Administer one or two of the following drugs 40 to 60 minutes prior to the procedure to mitigate the effects of aspiration if it occurs: a nonparticulate antacid (eg, sodium citrate 30 mL orally), a histamine-2 receptor antagonist (eg, ranitidine 50 mg intravenously), a proton pump inhibitor (eg, omeprazole 20 mg orally), or prokinetic drug (eg, metoclopramide 10 mg intravenously) Intubate the patient, especially at week 25 of gestation and beyond Intubation is the least preferred option for managing aspiration risk because of its associated morbidity, especially during pregnancy due to weight gain, edema, and hypervascularity; even minor trauma can lead to profuse bleeding. Intubation of pregnant patients is discussed separately. The teratogenic risks and postnatal effects of ECT are discussed separately.

EEG recording leads in ECT

Electroencephalogram (EEG) electrode placement. Reprinted with permission from: Handbook of ECT (Copyright 1997). American Psychiatric Publishing, Inc. Electrode placement for electroconvulsive therapy (ECT)

INDICATIONS FOR ECT The primary indication for ECT is for the treatment of major depression that is refractory to antidepressant medications [1]. Indications listed in the American Psychiatric Association guideline for the treatment of patients with major depressive disorder include psychotic depression, catatonic stupor, severe suicidality, food refusal leading to nutritional compromise, and pregnancy and other situations where a rapid antidepressant response is required (table 1) [2]. The report also recommends ECT for patients who have

previously shown a positive response to it and for those who have medical conditions that prevent the use of antidepressant medications. The Canadian Psychiatric Association clinical guidelines for the treatment of depressive disorders suggest similar indications [3]. Other psychiatric conditions for which ECT is effective include bipolar disorder, mania, and atypical psychosis [5]. ECT also may be effective or have application in patients with organic delusional disorder, organic mood disorder, obsessive compulsive disorder, catatonia secondary to medical conditions, neuroleptic malignant syndrome, neuroleptic-induced Parkinsonism, and neuroleptic-induced tardive dyskinesia [5,6]. TECHNIQUE AND ANESTHESIA ECT is usually administered two or three times per week for a total of 6 to 12 treatments. The treatment typically causes a 30 to 60 second generalized tonic clonic seizure, an effect that is essential to the success of ECT. The patient is preoxygenated with supplemental oxygen (2 L/min) via nasal cannula while the procedure is being set up. Prophylactic beta blockers may be administered immediately before or during ECT to blunt the hypertensive, tachycardic response to the seizure. The anesthesia of choice is methohexital. Other induction agents include propofol, thiopental, etomidate, and ketamine. Skeletal muscle relaxation is used during ECT to minimize the motor seizure and prevent musculoskeletal injury. The standard agent is succinylcholine via intravenous infusion. MORBIDITY AND MORTALITY Clinicians should be aware of certain potential side effects or complications of ECT. (See "Overview of electroconvulsive therapy (ECT) for adults", section on 'Adverse effects'.) The American Psychiatric Association lists the following conditions as associated with increased risk [7]: Unstable or severe cardiovascular disease Space-occupying intracranial lesion with evidence of elevated intracranial pressure Recent cerebral hemorrhage or stroke Bleeding or otherwise unstable vascular aneurysm Severe pulmonary condition ASA (American Society of Anesthesiologists) Class 4 or 5 (table 2) Mortality ECT is one of the safest procedures performed under general anesthesia. With modern anesthetic technique, the rate is sufficiently low and the potential life saving benefit is compelling enough that absolute contraindications to treatment no longer exist.

Cardiovascular effects A 15- to 20-second parasympathetic discharge occurs during the procedure as the patient enters the tonic phase of seizure. This can lead to arrhythmias including bradycardia with or without hypotension, atrial arrhythmias, premature atrial and ventricular contractions, atrioventricular block, and asystole. Asystole can occur with the first treatment or at any time later in a patient's course [8]; patients are at higher risk with longer periods of

subconvulsive seizures [9]. In one study of older adult patients, 66 percent had asystole lasting greater than 5 seconds with no lasting complications [10]. A history of hypertension or evidence of ischemia on electrocardiogram (ECG) did not predict asystole, nor did current use of calcium channel blockers, nitroglycerin, angiotensin converting enzyme (ACE) inhibitors, diuretics, or psychiatric medications (beta blockers were not included). Interestingly, patients with heart block and/or rhythm abnormalities were less likely to develop asystole (54 versus 16 percent). The clonic phase of the seizure then leads to a catecholamine surge that causes tachycardia and hypertension. The duration of tachycardia usually correlates with the length of the seizure itself [11]. These hemodynamic responses continue into the postictal period and usually resolve within 10 to 20 minutes of the seizure [12]. Occasionally patients have persistent hypertension that requires treatment. All patients are followed with ECG during the procedure; even healthy patients can have transient ECG changes. However, these changes are rarely significant, as illustrated by a study of 29 patients comprising a total of 80 treatments, in which no one had persistent T wave inversion, pathologic Q waves, or a demonstrable rise in cardiac enzymes at four and six hours (although some had a mild increase in creatine kinase [CPK], presumably from skeletal muscle) [13]. ECT can also cause a transient depression in the ejection fraction of healthy patients [14]. A study in 53 adults undergoing ECT found that seven developed new global left ventricular (LV) systolic dysfunction and eight developed regional wall motion abnormalities [15]. Among the 14 patients who developed global or regional abnormalities after the first ECT treatment, 13 had resolution of these abnormalities after the fourth ECT treatment (generally about one week later), and there were no short-term adverse events in any of the patients with LV dysfunction. Several studies have investigated the incidence of serious cardiac complications from ECT: In a retrospective study of 42 patients who had undergone ECT, 12 of the 17 patients with underlying cardiac disease had cardiac complications [16]. Most of these were benign and selflimited, including atrial and ventricular ectopy and nonsustained ventricular tachycardia. There was evidence of ischemia in two patients and one cardiac arrest of unknown etiology. Seventy percent of all complications, self-limited or not, occurred in patients with cardiovascular disease identified on the history, physical examination, or ECG, and all occurred in patients greater than 50 years of age. This study did not, however, systematically document the degree of preexisting cardiac disease. A prospective study of 40 patients with cardiac disease found that 55 percent had at least one complication with ECT [17]. These included minor complications, such as transient arrhythmias or ST segment changes, and major complications, such as persistent ECG changes accompanied by chest pain, asystole, or persistent arrhythmias. Only 7.5 percent in the control group without cardiac disease had a cardiac complication, and all were transient and minor.

A case-control study of 80 patients over the age of 50 found that the risk of major complications was 11.5 percent in patients at high risk for cardiovascular disease [18]. The patients were able to complete their course of therapy once the complications were treated. These findings suggest that the incidence of important cardiac complications are relatively rare and almost always occur in older patients and those with underlying cardiovascular disease. Furthermore, even patients at high risk for cardiac complications tolerate ECT well and, if complications occur and are treated, the vast majority can complete the treatment course. As an example, one study treated 80 patients, one-half of whom had preexisting cardiac disease (depressed ejection fraction, conduction disease, or frequent premature ventricular contractions) [17]. Among those with cardiac disease, eight had major complications including chest pain, arrhythmia, ischemia, and one myocardial infarction. Anesthesiologists and cardiologists treated the complications as they occurred and 36 of 38 patients went on to complete treatment, including the patient with the myocardial infarction. Similarly, a second study treated 53 patients and, of the 27 patients at risk, 31 percent required changes in medication or pretreatment with subsequent treatments because of complications [18]. Twenty-five out of 27 patients completed the course of ECT; there were no deaths. Despite these successful outcomes, it is not known how many of the patients in these studies were on beta blockers or other antihypertensive medication prior to treatment. Central nervous system and other effects Several cerebral effects occur with ECT, including increases in cerebral blood flow and intracranial pressure. Memory loss, disorientation, and delirium are the primary clinical manifestations. (See "Overview of electroconvulsive therapy (ECT) for adults", section on 'Adverse effects'.) ROLE OF THE MEDICAL CONSULTANT The preprocedure evaluation of patients before ECT is similar to the approach to a patient undergoing any procedure that requires general anesthesia: to identify medical issues that place the patient at risk, to propose strategies to reduce risk, and to treat complications after the procedure. Preprocedure evaluation A complete history and physical examination will help to identify pertinent risk factors [19]. We agree with the recommendation from the American Psychiatric Association that no specific laboratory tests are required in the pre-ECT evaluation. We recommend measurement of serum electrolytes only for patients taking diuretics or other medications that increase the likelihood of an abnormality and for patients with established renal disease or congestive heart failure. We suggest an electrocardiogram in patients over the age of 50 years, since most cardiac complications occur in older patients. There are no indications for other routine laboratory studies. The history should include a review of previous difficulties with anesthesia or ECT. As mentioned above, there are no absolute contraindications to ECT and most patients can have the procedure without serious complication, but clinicians should seek risk factors that may require

intervention or management. Important considerations include risk factors for cardiac ischemia or arrhythmia, heart failure, and the presence of brain tumors or other neurosurgical issues. A history of skull fractures should be determined, as this may affect electrode placement. The history should also include a medication history, to specifically inquire about use of herbal medications. In one study, 54 percent of outpatients with psychiatric conditions used alternative medication in addition to routine pharmacotherapy [20]. Common alternative medications, including Ginkgo biloba, ginseng, St. John's wort, valerian, and kava kava, have CNS effects which might interfere with ECT. The treating psychiatrist should be made aware of a patient's herbal supplement use, to determine if the herbal products should be tapered prior to initiating ECT [21]. In addition, theophylline has been reported to cause status epilepticus after ECT and so should be tapered prior to treatment [22]. The treating psychiatrist may also choose to taper or discontinue antidepressants and/or other psychotropics prior to treatment. A summary of management strategies for preexisting medical conditions is shown in a table (table 3). Strategies to reduce the risk of cardiac complications The 2007 American Heart Association /American College of Cardiology (AHA-ACC) guideline update for noncardiac surgery assigns procedure-related risk to several types of surgery [23]. Though not explicitly mentioned in the AHA-ACC guidelines, ECT can be treated like a low-risk procedure because it is usually well tolerated even in those at risk, the duration of hemodynamic changes is brief, and the mortality rate is low [17,18,24]. As a result, it is difficult to prove a beneficial effect of any intervention to reduce cardiovascular risk. Nevertheless, a cardiac risk assessment is warranted preoperatively. Barring major clinical predictors of coronary risk (unstable angina, decompensated heart failure, severe valvular disease, malignant arrhythmias), most patients can undergo ECT with appropriate medical management [25]. Some have suggested that sedentary patients with intermediate risk factors should undergo further noninvasive risk stratification [26]. However, we feel that this population can proceed without further evaluation given the low morbidity of ECT and the potential use of prophylactic beta blockers. Postprocedure hemodynamic changes Short acting beta blockers are used to quickly treat persistent or severe tachycardia and hypertension after ECT. Other agents that have been useful for the treatment of postprocedure hypertension are intravenous nitroglycerin [27], nicardipine [28], and clonidine [12,25]. Prophylactic beta blockers Short acting beta blockers, such as intravenous esmolol or labetalol, can be used to prevent transient and persistent hypertension and tachycardia [27-33];

glycopyrrolate is administered to prevent bradycardia in these individuals. Several authors have raised concern regarding widespread use of beta blockers because these drugs have been implicated in cases of prolonged asystole [9,34,35]. However, in these reported cases, atropine was not used and in some cases patients had subconvulsive treatments that can also cause bradycardia. While it is clear that treatment of postprocedure hemodynamic changes is indicated, use of prophylactic beta blockers to reduce risk for patients undergoing ECT is controversial. The overall risk of cardiovascular complications due to ECT is low and no studies have documented reduced cardiac complications with the use of short acting beta blockers before ECT. The 2001 American Psychiatric Association guidelines list three considerations for the use of prophylactic short acting beta blockers, that also are relevant in patients taking long acting beta blockers [7]: Beta blockers may increase the risk of asystole (though this appears to be less of a risk if atropine is used [10]). The absolute rise in the heart rate and blood pressure with ECT is no greater in those with baseline elevated blood pressure or heart rate [36,37]. There is a theoretical risk that excessive blunting of the hemodynamic response with beta blockers may reduce the necessary supply of oxygen to the brain to protect against decreased seizure length and cognitive side effects. In addition, there is some evidence that beta blockers might result in a decrease in seizure length and potentially decrease the efficacy of ECT [33,38], although this is not found in all studies [39,40]. The American Psychiatric Association chose not to make a specific recommendation with regard to using prophylactic beta blockers but does recommend that patients with unstable hypertension be stabilized prior to starting ECT [7]. We do not recommend routine use of beta blocker in low risk patients, given the theoretical risks and unclear benefit. We suggest the use of prophylactic short-acting intravenous beta blockers for patients at very high risk of complications from transient hypertension (eg, intracranial aneurysm, unstable angina or recent myocardial infarction) [19]. Collaboration with the treating psychiatrist, anesthesiologist, and cardiology consultation is beneficial in such patients. Typical bolus doses of labetalol are 5 to 20 mg IV, and for esmolol, 10 to 50 mg IV. The most authoritative study of the use of beta blockers during ECT found that labetalol produced dosedependent reduction in HR and rate pressure product (RPP) using 5 and 10 mg doses, compared to placebo [41]. It is not necessary to prevent the transient hypertension and tachycardia associated with the seizure in most patients; the main goal of the use of these agents is to reduce the risk of myocardial ischemia (a possible result of the increased oxygen demand associated with tachycardia) in patients with coronary artery disease. Other agents, such as calcium

channel blockers or nitroglycerine, are sometimes given during ECT [42]. Other prophylactic medications Medications other than beta blockers have been used to try to prevent postprocedure hemodynamic changes. As an example, a small randomized trial found that prophylactic intravenous nicardipine was effective in a dose-dependent fashion at minimizing the acute hemodynamic response to ECT without shortening the duration of seizure [43]. Coexisting cardiac disease Hypertension Patients should receive their routine antihypertensive (other than diuretics) with a small sip of water approximately two hours before ECT. Diuretics should not be given, because it is better for the bladder to be empty during the procedure to prevent the patient from soiling himself as a result of the seizure. Coronary heart disease Antianginals such as nitrates and beta blockers should be continued in patients with documented coronary heart disease who are already taking them. Patients taking long acting beta blockers should receive atropine or glycopyrrolate with induction given the potential increase in the risk of bradycardia and a small study that showed that atropine may be protective [10]. If ECG changes or chest pain occur during the treatment period, treatments should be postponed until the patient is evaluated and treated for potential ischemia [25,44]. Heart failure and valvular disease ECT should be delayed in patients with decompensated heart failure or significant valvular disease, pending cardiology consultation and completion of a thorough evaluation and optimization of cardiac status. One report of ten patients who completed ECT (total of 144 ECT sessions) with severe aortic stenosis (valve area 0.7 to 1.0 cm2) noted good control of blood pressure and heart rate with continuation of chronic antihypertensive medications and/or addition of a short-acting beta blocker [45]. All patients were asymptomatic and tolerated ECT. There were no deaths attributable to ECT or aortic stenosis. Two patients had drops in blood pressure requiring intervention, and seven patients required intravenous medication to control systolic blood pressure >180 mm Hg. In patients with severe aortic stenosis, clinicians should strive for tight control of blood pressure and heart rate with attention to avoiding excessive preload or afterload reduction. Little data are available for management of patients with less severe valvular disease, though ECT can be performed safely, with appropriate precautions, in most patients with underlying cardiac conditions [46]. In patients with a remote history of heart failure or compensated disease, we recommend a baseline echocardiogram, if not recently performed, to assist with periprocedure management. In those with compensated systolic dysfunction, one may simply continue diuretics and vasodilator therapy and minimize volume overload. In patients with diastolic dysfunction,

control of blood pressure should limit the occurrence of flash pulmonary edema. This group will benefit from prompt treatment of post-seizure hypertension if it develops. Pacemakers and implantable defibrillators Patients with pacemakers and automatic implantable cardiac defibrillators (AICDs) can safely undergo ECT. As an example, in a case series of ECT in 26 patients with pacemakers and three patients with AICDs there was only one serious cardiac event, an episode of supraventricular tachycardia [47]. The clinical team should be prepared to deactivate pacemakers with a magnet if any aberrant signals occur [25]. A cardiologist should deactivate AICDs before induction and reactivate the device after the seizure is complete to avoid excessive charge. The patient should have continuous ECG monitoring while the defibrillator is deactivated [48]. Coexisting neurologic and neurosurgical disease Brain tumors Early recommendations considered brain tumors and other space occupying lesions to be absolute contraindications to ECT [49]. This concern was based upon the observation that ECT raises cerebral blood flow, which in the presence of a brain tumor could translate into an increase in intracranial pressure (ICP) and neurologic deterioration. These concerns were illustrated in a review of 35 patients with brain tumors undergoing ECT, 74 percent of whom had major adverse neurologic and cognitive side effects, with a one-month mortality of 28 percent [50]. Notable characteristics of patients who suffered neurologic deterioration included the presence of a depressive illness, no previous psychiatric illness or ECT, headache, and the presence of even soft neurologic findings on exam. However, this study may have had a selection bias given that only one patient was known to have a tumor before treatment. Subsequently, one group reviewed 10 cases of successful treatment [51-54]. Eight patients had meningiomas in differing locations, and two patients had metastatic breast cancer. No one had abnormal neurologic examinations or evidence of increased ICP at baseline, and four had normal spinal fluid measurements. There were no adverse events other than a prolonged seizure after one treatment. Most recently a case report described successful treatment of a patient with primary brain cancer and documented elevated ICP [55]. He was treated with steroids and short acting beta blockers to minimize edema and elevations in blood pressure during the clonic phase of seizure. Based on this information, ECT is probably safe in patients with brain tumors as long as there is no evidence of elevated ICP. Given that the available safety data are in the form of case series or reports, we agree with the American Psychiatric Association that the decision should be made on a case by case basis with the involvement of neurologic and possibly neurosurgical consultants [7]. Evaluation should consist of a detailed history and physical examination to seek

evidence suggestive of elevated ICP, such as headache, papilledema, or abnormal neurologic examination. Any such evidence should prompt further investigation such as head CT or MRI. Although data are limited, providers have successfully used steroids to minimize edema [54,55], and neurologic consultants may consider using them in patients in whom elevated ICP cannot be ruled out [56,57]. Stroke Patients who have suffered a stroke have a high rate of depression. In a study of 14 patients with completed strokes undergoing ECT one month or more after the event, none had deleterious neurologic sequelae [58]. Similarly, a second study of 24 patients with a history of stroke found no difference in the efficacy or cardiac complication rate compared with controls [59]. Rates of delirium were similar in both groups, although within the study group all patients with delirium had their strokes within the preceding year. Adequate blood pressure control is important for such patients. Dementia Reports exist of successful ECT in depressed patients with dementia and organic brain disease, with efficacy rates similar to that of nondemented patients [60,61]. In a review of 135 patients with organic and depressive dementia, 21 percent developed delirium or cognitive and memory deficits; all but one had cleared by time of discharge from the hospital [60]. The severity of delirium correlates with the degree of underlying dementia or organic brain syndrome, but is transient and does not interfere with treatment [61]. The clinical staff should be aware that delirium is a potential side effect in demented patients. These patients can often be managed with reassurance and alterations in level of supervision [62]. Neuromuscular disease In patients with neuromuscular disease, particularly post-polio syndrome, depolarizing muscle relaxants such as succinylcholine may lead to severe hyperkalemia and circulatory collapse [63,64]. Thus, depolarizing muscle relaxants should be avoided. Patients with neuromuscular disease may receive short-acting, non-depolarizing agents for neuromuscular blockade [65]. Epilepsy Use of ECT and anticonvulsants in patients with epilepsy is discussed separately. Diabetes There is no clear evidence of an effect of ECT on blood sugar control in diabetic patients. In a study of 19 patients with insulin-requiring type II DM, ECT itself did not lead to significant acute changes in blood sugar [66]. Individual patients who developed hyperglycemia or hypoglycemia after ECT were found to have changed their behaviors (eating or activity) in response to resolving depression, and these behavioral changes were felt to be responsible for the change in blood sugar control. Patients generally do not take anything by mouth on the morning of ECT. As a general rule, we recommend holding oral hypoglycemic agents on the morning of the procedure. For insulinrequiring diabetics, on the morning of the procedure we recommend giving half the usual long-

acting insulin dose and withholding short-acting insulin. A more detailed discussion of these issues can be found elsewhere. (See "Perioperative management of diabetes mellitus".) Anticoagulation The safety of ECT in patients who are anticoagulated is controversial because of concerns about a possible increase in the risk of intracerebral hemorrhage [67]. In a case series of 35 patients receiving long-term anticoagulation with warfarin who underwent 284 ECT treatments, no major adverse effects occurred [68]. The INR on the day of ECT was below subtherapeutic (below 2.0) 36 percent of the time and supratherapeutic (above 3.5) only 3 percent of the time. These data suggest that if there is an increased risk of ECT in patients appropriately anticoagulated with warfarin, that risk is small. Pregnancy The treatment of psychiatric conditions in pregnancy poses challenges, as psychotropic medications may have significant side effects in both mother and fetus. However, ECT is generally thought be safe in pregnant patients by the American Psychiatric Association and the American College of Obstetricians and Gynecologists [7,69]. The safety of ECT and modifications in technique for treating pregnant patients are discussed separately. As a pregnant patient considers ECT, the informed consent and evaluation should be performed with an obstetrician and an anesthesiologist.

Ablation Neurosurgery INTRODUCTION Many patients with unipolar major depression do not respond to standard treatment with pharmacotherapy and psychotherapy [1,2], and are thus candidates for neuromodulation procedures [3-8]. Electroconvulsive therapy (ECT) is the oldest neurostimulation procedure (it also predates all antidepressant drugs), and is the most effective intervention for unipolar major depression [3,4,7]. In addition to ECT, other noninvasive neurostimulatory therapies include magnetic seizure therapy, focal electrically administered seizure therapy, transcranial magnetic stimulation (TMS), transcranial direct current stimulation, transcranial low voltage pulsed electromagnetic fields, and cranial electrical stimulation [9]. Invasive neuromodulation interventions (which require surgery and have generally been studied in more treatment-refractory patients) include vagus nerve stimulation (VNS), deep brain stimulation, direct cortical stimulation, and ablative neurosurgery. This topic provides an overview of noninvasive and invasive neuromodulation therapies for unipolar major depression. The initial treatment of depression and management of treatment resistant depression are discussed separately. (See "Initial treatment of depression in adults".) NONINVASIVE NEUROMODULATION THERAPIES Noninvasive neuromodulation

procedures use an electric current or magnetic field to stimulate the central nervous system [8]. Convulsive therapies Noninvasive neurostimulation treatments for major depression include convulsive therapies: Electroconvulsive therapy (ECT) Magnetic seizure therapy Focal electrically administered seizure therapy Electroconvulsive therapy (ECT) Electroconvulsive therapy (ECT) is a clinically available approach that uses an electric current that passes between two electrodes placed against the scalp to induce a generalized cerebral seizure while the patient is under general anesthesia. A course of ECT involves a series of treatments that are delivered over several days to weeks. ECT is superior to pharmacotherapy for unipolar major depression based upon meta-analyses of randomized trials [10,11], and is generally considered the most efficacious treatment for depression [12]. However, ECT is associated with safety risks, adverse effects, logistical constraints, and patient refusal, and relapse rates following remission are high, especially in patients with treatment resistant depression [13,14]. INVASIVE/SURGICAL NEUROMODULATION THERAPIES Invasive neuromodulation therapies for treating unipolar major depression include: Vagus nerve stimulation Deep brain stimulation Direct cortical stimulation Ablative neurosurgery

Invasive neuromodulation interventions require surgery and have generally been studied in patients with chronic, treatment-refractory, debilitating depression, because most clinicians view invasive interventions as riskier than the noninvasive techniques described above. Among the surgical approaches, vagus nerve stimulation, deep brain stimulation, and direct cortical stimulation are generally: Reversible Hardware can be removed Revisable Stimulating electrodes can be moved to optimize response Adjustable Stimulation parameters can be modified to optimize response However, ablative surgery does not involve: In dwelling metal hardware and contraindications to magnetic resonance imaging and metal detectors

 Surgical follow-up to replace batteries or pulse generators every few years or to revise dysfunctional systems

Ablative neurosurgery Ablative neurosurgery for intractable major depression is a clinically available but rarely used approach in which a lesion is made in limbic or paralimbic structures. Early ablative procedures for psychiatric illness, such as the prefrontal leucotomy [59], have been supplanted by stereotactic neurosurgical techniques that allow for more focal lesions with fewer side effects, including [60-62]: Anterior capsulotomy Lesion in the anterior limb of the internal capsule Anterior cingulotomy Lesion in the dorsal anterior cingulate Subcaudate tractotomy Lesion in thalamocortical white matter tracts inferior to the anterior striatum Limbic leucotomy Combines cingulotomy with subcaudate tractotomy

Unipolar major depression often does not respond to standard treatment with pharmacotherapy, psychotherapy, and noninvasive neuromodulation interventions such as electroconvulsive therapy (ECT) and transcranial magnetic stimulation (TMS). This has led to investigation of invasive treatments that include: Most studies of surgical neurostimulation interventions have maintained patients on their presurgical antidepressant medications and psychotherapies after surgery [3,4]. Assessment Candidates for surgical treatment of refractory unipolar major depression should be evaluated to confirm the diagnosis and whether surgery can be performed safely. The assessment includes a psychiatric history and mental status examination, with emphasis upon current depressive (table 1) and psychotic symptoms, comorbid disorders (eg, substance use and personality disorders), neuropsychological functioning, length of the current depressive episode, types and number of failed treatments during the present episode, psychosocial functioning, and ability to provide informed consent, as well as the number, length, and treatment history of prior depressive episodes. In addition, a general medical history and physical examination is performed, as well as laboratory tests and neuroimaging studies that are guided by the history, examination, and possibility of surgery [5-7]. The medical work-up should emphasize preexisting neurologic disease (eg, epilepsy, intracranial masses, and vascular abnormalities). Craniotomy is typically used to perform these procedures, although a gamma knife can be used

for capsulotomy because of its particularly small lesion volume [55]. Cingulotomy has been used most often, with over 500 reported cases [36]. Indication There are no established criteria (including number and type of failed treatments) that need to be met before a patient with unipolar major depression is considered appropriate for ablative surgery; this largely depends upon the clinical judgment of the evaluating psychiatrist and neurosurgeon. Candidates include patients with chronic, severe, and incapacitating symptoms that are refractory to [1,2,56,57]: Pharmacotherapy Monotherapy with several (eg, four) different antidepressant classes at therapeutic doses for several weeks (eg, four to eight) Multiple (eg, three) trials of an antidepressant plus an established adjunctive drug (eg, second-generation antipsychotic, lithium, and triiodothyronine) Adjunctive psychotherapy such as cognitive-behavioral therapy or interpersonal psychotherapy Electroconvulsive therapy (ECT) Contraindications to ablative surgery for unipolar major depression include [1,56,57]: Severe personality disorders Comorbid substance use disorders Suicidal ideation or behavior Chronic, poorly controlled general medical conditions Previously diagnosed intracranial masses Intracerebral vascular abnormalities Pregnancy Institutions offering ablative neurosurgery for treatment refractory depression often have a multidisciplinary committee to review referrals and ensure that the procedure is indicated [1,2]. In many countries, use of the procedure is controlled by legislation and overseen by public agencies [2,51,56].

Before undertaking ablative neurosurgery, clinicians and patients with major depression should note that the course of illness is such that recovery can occur after lengthy periods of illness. Response There are no randomized trials for ablative procedures in treatment refractory depression. Based upon prospective observational studies, response (improvement from baseline on the depression rating scale 50 percent) occurs in approximately 30 to 60 percent of patients; the evidence includes a study of: 33 patients treated with anterior cingulotomy or limbic leucotomy (mean duration of

follow-up 30 months); response occurred in 33 percent [1] 8 patients treated with anterior cingulotomy (duration of follow-up 12 months); response occurred in 63 percent [57] 20 patients treated with anterior capsulotomy (mean duration of follow-up 7 years); response occurred in 50 percent [2] In a retrospective study, 23 patients treated for major depression with subcaudate tractotomy or limbic leucotomy were interviewed, with a mean duration of follow-up of 14 years; recovery occurred in 22 percent and significant improvement in 48 percent [56]. However, 29 other patients were alive and lost to follow-up, and among 24 patients who were deceased at follow-up, 25 percent had committed suicide. Improvement of major depression with ablative surgery generally occurs within days to weeks of the procedure [56], but the effect may be delayed for months [1,58]. Insufficient or impersistent improvement may lead patients to undergo a second operation to extend the lesions created in the first operation or to make lesions in an additional site [1,2,56]. Risks and side effects For patients with treatment refractory depression, observational studies suggest that ablative surgery is associated with: Nonspecific neurosurgical risks [1,56]: Intracranial bleeding Infection Anesthesia complications Delirium Risks specific to the lesion [1,2,19,56]: Epilepsy Impaired cognition (including executive functioning, memory, set shifting, and verbal fluency) Personality changes (eg, impulsivity, disinhibition, and amotivation) Weight gain Urinary incontinence (typically transient) However, some studies found that neurosurgery did not adversely affect neuropsychological function [2,57]. Mechanism of action The rationale for using ablative procedures is based upon a model that conceptualizes refractory major depression as a problem in neural networks (circuits) that regulate mood, rather than an abnormality in a single neuroanatomical structure or neurotransmitter system [1,46]. It is hypothesized that abnormal communication between grey matter brain regions involved in the pathophysiology of major depression can be interrupted by severing their white matter connections.