Print Friendly 

In Session with Charles H. Kellner, MD: Current Developments
in Electroconvulsive Therapy

Charles H. Kellner, MD, interviewed by Norman Sussman, MD

Primary Psychiatry. 2007;14(3):34-37

This interview was conducted by Norman Sussman, MD, on December 15, 2006.

Disclosure: Dr. Kellner receives grant support from the National Center for Complementary and Alternative Medicine and the National Institute of Mental Health.

This interview is also available as an audio PsychCastTM at

If you have any questions and/or comments regarding the interview with Charles H. Kellner, MD, please submit a “Letter to the Editor” to Norman Sussman, MD, at

Dr. Kellner is professor and chair of the Department of Psychiatry, and assistant dean for clinical research at the University of Medicine and Dentistry of New Jersey in Newark. He is also co-director of the electroconvulsive therapy (ECT) service at The University Hospital in Newark with George Petrides, MD, and was editor of The Journal of ECT from 1994–2004. Dr. Kellner’s ongoing research includes studies comparing different ECT types and methods for optimizing maintenance ECT.


What typical procedures are involved in electroconvulsive therapy (ECT)?

Performed in a modern hospital setting, ECT is a series of treatments in which patients receive general anesthesia and a muscle relaxant followed by a small amount of electrical current to induce a seizure that typically lasts 1 minute. Each procedure takes approximately 15 minutes. The patient is under anesthesia and feels no pain or discomfort.

How many treatments are typically needed before patients show a response?

While there is some variability, most patients with major depressive disorder (MDD) and other depressive disorders will typically remit with an acute ECT course of 6–10 treatments. Some patients achieve remission with fewer than six treatments and some patients will need more than 10 treatments.

From the early days of its application, clinicians saw remarkable results with ECT. It continues to be the most effective treatment available for MDD. However, we are still unsure as to why ECT works.

What are current indications for ECT versus drug therapy?

The most important use of ECT is in patients with MDD who have not responded to medication. Approximately 80% of all patients who receive ECT have very severe depression, either in the context of unipolar or bipolar disorder, which has not responded to several trials of medication.

In some situations, a patient has such a severe case of MDD that medication is not used. Rather, ECT is indicated as a first-line treatment. Such patients would include those who are severely suicidal, extremely psychotic, extremely agitated, or so depressed that self-care is declining and impairing their health through dehydration or malnourishment. Thus, there are some acute situations where ECT is a primary treatment. However, the majority of patients will have received several medication trials. If those medications are not adequately effective in relieving MDD symptoms, ECT is an appropriate choice.

In addition to its antidepressant effects, ECT acts as a mood stabilizer and is effective in both poles of bipolar disorder. Although it is also an acute anti-manic treatment, ECT is not used much for that purpose because most manic patients do respond to medication. For the small subset of patients who do not respond to medication, ECT is an effective anti-manic treatment.

ECT also has a limited but critical role in the treatment of schizophrenia. While ECT does not treat the underlying illness, it treats the positive symptoms of schizophrenia. Thus, ECT can be helpful in treating hallucinations, delusions, and agitation, particularly early in the course of schizophrenia, and should not be overlooked as a treatment for that illness.

In addition, we have found that patients with comorbid MDD and Parkinson’s disease have had beneficial results after ECT, both in terms of improvement in the motor systems of Parkinson’s disease as well as depression. ECT is also a good anti-catatonia treatment, whether catatonia occurs in the context of bipolar disorder, schizophrenia, or, more rarely, other medical illnesses. ECT has the potential to bring a patient out of a catatonic state.

Does the memory impairment acutely associated with ECT persist over time?

ECT causes three types of memory impairment: an acute-confusional state, anterograde amnesia, and retrograde amnesia. The acute-confusional state is characterized by patient confusion due to the seizure and general anesthesia. That impairment typically lasts 20–30 minutes and subsides by the time patients leave the hospital. Patients are brought to a recovery room in order to wake up in an unstimulating environment because of this impairment. Anterograde amnesia is the disruption of memory function during the course of ECT treatment. If, for example, patients are receiving ECT three times a week for 3 weeks of treatment, they will have difficulty remembering and recording memories during that time. Anterograde amnesia dissipates over approximately 1–2 weeks after the end of the ECT course with memory functioning returning to normal. For that reason, patients cannot drive for approximately 1 week after their course of ECT, and certainly not during ECT treatment. Retrograde amnesia is the erasure of a portion of memory around the time of the treatment course, and is probably the most anxiety-producing type of memory problem caused by ECT. It is the only type of memory impairment that can have a long duration. Many memories can return, but some can be erased permanently.

There are reports of more extensive memory problems with ECT and, in rare cases, ECT has the theoretical potential to cause more severe memory impairment.1 Memory impairment was more common in the past when ECT techniques were not as sophisticated as they are currently. Many refinements of ECT technique have occurred over the last 25 years and have served to minimize the amount of retrograde memory impairment caused by ECT.

Is ECT ever indicated for someone who has previously failed a course of ECT?

It is possible that a patient who did not respond during a particular episode could respond in a subsequent episode. Clinicians should carefully review how ECT was performed and patient circumstances. In situations where patients may have had adverse effects from ECT, clinicians should be careful about using the treatment again, particularly if patients have had more cognitive effects than predicted.

Sometimes, ECT is not effective because patients have not received a sufficient number of treatments. In research studies, 10 treatments are conducted before declaring that a patient is a nonresponder.2 Clinicians should take all patient information into account, but 10–12 or more treatments is not unusual. If a patient is not responding in the middle of an ECT course, the practitioner should review the patient diagnosis and ensure that patients are not taking any concomitant medications that may be interfering with ECT efficacy. Clinicians should ensure that the technical parameters of treatment are optimized for each patient.

Should antidepressants or mood stabilizers be stopped during treatment?

Years ago, clinicians used to stop all medications while patients received ECT. However, most antidepressants can be safely continued during ECT. There is some developing evidence that medication may be beneficial in terms of efficacy and in preventing relapse after ECT.3 In addition, commonly-used antidepressants are now much less cardiotoxic than older tricyclic antidepressants, and we often co-administer them with ECT.

Some medications are contraindicated in ECT. Anticonvulsants interfere with seizure production, which is the therapeutic agent of ECT. Antiepileptic mood stabilizers should be tapered and stopped before ECT. Benzodiazepines should only be given in small doses concurrently with ECT because they are potently anticonvulsant. Monoamine oxidase inhibitors (MAOIs)—a class of antidepressants used in the 1960s and 70s that is experiencing a resurgence—can also be safely co-administered with ECT. There are two methods of working with MAOIs when patients receive ECT. It is most conservative to stop MAOIs for 10–14 days prior to treatment. However, if patients are receiving benefit, MAOIs can often be safely continued during the course of ECT. The anesthesiologist and psychiatrist should be aware of potential problems and have the ability to treat the problems if they do occur. Antipsychotics, which are often co-administered in patients with psychotic depression, schizophrenia, and anxiety, can be safely co-administered with ECT as well.

Does ECT ever convert medication nonresponders to responders?

We do not know. It is theoretically possible, but that is a very hard occurrence to study. Another possibility is that patients can partially respond to antidepressants and ECT can help them achieve remission. Continuing antidepressant use has beneficial effect in preventing relapse once patients achieve remission.

Are there any strategies for extending the benefit of an ECT treatment course?

There are two shortcomings of ECT. The first is memory impairment. The second is that there is a high relapse rate in the first 6–12 months if clinicians do not aggressively treat patients who have had successful ECT. Researchers and clinicians are addressing that shortcoming in several ways. First, we are more prone to utilize tapered courses of ECT, also referred to as continuation ECT. In the past, patients may have received treatments three times a week and then stopped abruptly. Now, patients are treated until symptom remission and receive treatments on a lengthened schedule over the next several weeks. A patient may receive treatments three times a week and return for additional treatments at 1-week intervals and later at 2-week intervals. In addition, we are more liberal in co-prescribing antidepressants during the acute course of ECT. This method may have the benefit of ensuring that patients have a therapeutic level of antidepressants when ECT is stopped. In the past, if patients had an acute course of ECT on no medication and started antidepressants after the treatment course, they may not have a therapeutic blood level of the antidepressant for another 1–2 weeks, which would leave them at risk for relapse.

In addition, if a patient has a serious history of recurrent episodes of mood disorder, they can be considered for maintenance ECT, which is an extended course of ECT past the continuation period. The patient would receive a single treatment on an outpatient basis approximately one time per month, although there is flexibility in that schedule. The goal is for patients to receive ECT, even if they are well, about once a month as a prophylactic treatment to prevent a subsequent episode of mood disorder.

Is there evidence that repeated ECT causes brain damage?

There is much evidence that ECT does no structural damage to the brain.4 There are a number of well-designed studies that show ECT does not cause brain damage and numerous reports of patients who have received a large number of treatments over their lifetime and have suffered no significant problems due to ECT.5 In fact, there is evidence that ECT may cause the release of trophic chemicals from neurons that may actually protect the brain. This effect has been shown in animal studies in which ECT induces sprouting of neurons in the hippocampus and may cause increased proliferation of neurons.6 Rather than cause brain damage, there is evidence that ECT may reverse some of the damaging effects of serious psychiatric illness. Several magnetic resonance imaging studies have demonstrated that there is no evidence of structural brain damage due to ECT.7

What are the prevailing theories as to how ECT works?

ECT is a powerful treatment that affects multiple brain systems. While researchers understand the action of ECT on the brain, because it effects many brain regions simultaneously, we have not yet been able to determine which effect or multiple effects causes the antidepressant or antipsychotic effects of ECT. There are theories suggesting that ECT and antidepressant medication action act similarly, each releasing neurotransmitters that are deficient in depressed states. However, ECT acts on a more powerful scale.

Over the last 20 years, we have observed that, parodoxically, ECT is an anticonvulsant treatment. When clinicians induce a series of seizures, they are teaching the brain to recruit anticonvulsant mechanisms to end the seizures. Researchers in psychopharmacology know that one of the most important advances in the treatment of mood disorders in the last 20 years was the introduction of the anticonvulsant mood stabilizers.

There is reason to believe that the anticonvulsant property of ECT may be related to its antidepressant effect. Other theories include one called the neuroendocrine theory, which states that ECT causes the release of peptides that are beneficial in restoring appropriate mood. There is also evidence that ECT corrects the main neuroendocrine abnormality in MDD, which is disregulation of the hypothalamic pituitary adrenal axis. This result was shown in work with the dexamethasone suppression test. In these studies, patients successfully treated with ECT showed normalization of their dexamethasone suppression test.8

There are also many attempts to refine ECT, prevent relapse, and uncover the mechanism of ECT action. Regarding relapse prevention, there are studies now looking at combining maintenance ECT with pharmacotherapy to prevent relapse.

There are ongoing studies examining the best technical methods to perform ECT.9 Three techniques in performing ECT are bilateral (also called bitemporal) electrode placement, right unilateral electrode placement, and the more recent bifrontal placement. Studies are looking at the relative efficacy and the cognitive side-effect profile of each of these three electrode placements. These are attempts to make treatment as effective as possible while limiting cognitive side effects.

Should clinicians begin ECT treatment sooner than previously thought or utilize implanted vagus nerve stimulation (VNS)?

In many situations, it is appropriate to consider ECT earlier rather than later. Modern ECT is a benign procedure. Thus, I believe it is a clinical mistake to allow a patient to remain severely ill and subjected to many antidepressant trials before offering ECT. The application of ECT should be determined by the severity of the patient’s illness. For those patients who are most severely symptomatic with typical MDD symptoms, clinicians should consider ECT among the treatment options. Unfortunately, many clinicians are reluctant to consider ECT sooner. This is a prejudice that should be overcome.

More evidence is needed on the efficacy of VNS as a treatment before it is given widespread use. Data on VNS efficacy are emerging. While evidence is encouraging, it is premature to conclude that it will be an important treatment in our field. PP


1. Sackeim HA. Memory and ECT: from polarization to reconciliation. J ECT. 2000;16(2):87-96.

2. Kellner CH, Knapp RG, Petrides G, et al. 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(12):1337-1344.

3. Sackeim HA, Haskett RF, Mulsant BH, et al. Continuation pharmacotherapy in the prevention of relapse following electroconvulsive therapy: a randomized controlled trial. JAMA. 2001;285(10):1299-1307.

4. Weiner RD. Does electroconvulsive therapy cause brain damage? Behav Brain Sci. 1984;7(1):1-22.

5. Devanand DP, Verma AK, Tirumalasetti F, Sackeim HA. Absence of cognitive impairment after more than 100 lifetime ECT treatments. Am J Psychiatry. 1991;148(7):929-932.

6. Duman RS, Vaidya VA. Molecular and cellular actions of chronic electroconvulsive seizures. J ECT. 1998;14(3):181-193.

7. Coffey CE, Weiner RD, Djang WT, et al. Brain anatomic effects of electroconvulsive therapy. A prospective magnetic resonance imaging study. Arch Gen Psychiatry. 1991;48(11):1013-1021.

8. Swartz CM, Saheba NC. Dose effect on dexamethasone suppression testing with ECT. Ann Clin Psychiatry. 1990;(2):183-188.

9. Kellner CH, Knapp R, Petrides G, et al. Bifrontal stimulation: where are we right now? J ECT. 2006;22(1):78.