Although conventional electroconvulsive therapy (ECT) with bitemporal (BT) electrode placement in depressed patients commonly achieves remission rates greater than 80%, fears of the occasional prolonged cognitive effects inhibit patients and practitioners in its use. New research in ECT regarding variations in electrode placements among BT, bifrontal (BF) and unilateral (RUL) electrode placements is finding improved efficacy and a reduction in adverse memory effects. These findings strengthen the guidelines for delivering the most effective courses of ECT treatment.
Bitemporal Electrode Placement
Bitemporal placement of the stimulating electrodes, one on each temple, is the conventional placement in ECT. This placement is widely used, effective and easy to administer. Bitemporal placement is used in the ongoing CORE Study Group, which is designed to assess the merits of different continuation treatments after a successful index course of ECT for patients with unipolar depression. The CORE Study Group is a collaboration of scientists at the Medical University of South Carolina in Charleston (Charles Kellner, M.D., Principal Investigator [PI]); the Mayo Clinic in Rochester, Minn. (Teresa Rummans, M.D., PI); Long Island Jewish (LIJ)-Hillside Medical Center in New York (Georgios Petrides, M.D., PI); and the University of Texas Medical Center at Dallas (Mustafa Husain, M.D., PI). The study began in February 1997 and is supported by grants from the National Institute of Mental Health.
The data are interim findings from the first phase of the study in which depressed patients were treated with BT ECT at 1.5 times the measured seizure threshold. After one week of remission, patients were randomly assigned to either continuation ECT or continuation pharmacotherapy with the combination of nortriptyline (Pamelor, Aventyl) and lithium, both sustained at minimum-measured, monitored blood serum levels, for six months.
The interim findings found that 217 (86%) of 253 enrollees completed the prescribed course of ECT in the first three years. Of these completers, 189 (87%) remitted. Remission is defined by the rigorous standard of a sustained reduction in the 21-item Hamilton Depression Rating Scale (HAM-D) greater than 60% and maximum scores of 10 or less for one week. Of all patients entered into the study, 75% remitted.
Sixty-seven of the patients (31%) were both psychotic and depressed and, of these, 64 (95%) achieved remitter status, a higher rate than among the nonpsychotic depressed (125/153, 83%). The psychotic patients responded faster and with greater reductions in HAM-D scores at all time points. The difference was evident after five ECT treatments and was sustained throughout the study.
The patients referred for ECT were severely ill. The average baseline HAM-D score was 37.8 for the psychotic and 33.8 for the nonpsychotic depressed patients. The baseline average HAM-D of 35 for all patients was reduced to 9.6. It took approximately seven ECT treatments for remission, an average of 16 days.
The older patients in this study recovered faster and to a greater degree than the younger patients. Of those between ages 18 and 45 years, the average baseline HAM-D was reduced to 12.7, while for those between 46 and 64 years and 65 and 85 years, the HAM-D scores fell to 8 and 8.5, respectively. The study found the antidepressant effect of BT to be robust among psychotic depressed and elderly depressed patients.
Bifrontal Electrode Placement
Bifrontal electrode placement -- each electrode on the brow over the outer canthus of the eye -- was re-introduced a decade ago in two Canadian studies that compared BF and BT at just over threshold energy dosing and RUL at 2.5 times the titrated seizure threshold (ST) (Lawson et al., 1990; Letemendia et al., 1993). When 40 depressed patients were assigned to one of three ECT electrode placement groups, they showed good therapeutic efficacy with BF, comparable to BT, with a sparing of the effects on verbal and nonverbal cognitive tests (Lawson et al., 1990). A follow-up study that assigned 59 patients to the same treatment groups reported greater benefits for BT and BF compared to RUL, with an advantage in adverse side effects for BF (Letemendia et al., 1993).
These results were independently confirmed in a study of 48 patients treated with either BF or BT (Bailine et al., 2000). Patients in both groups met remission criteria by the 12th treatment. Those treated with BF exhibited lesser effects on cognitive tests. A new replication study from the Canadian group found equivalent efficacy across the three placements, allowing the authors to conclude that treatment with BF electrode placement provides the best ratio of benefits to side effects and should be given at near threshold level to minimize cognitive effects (Delva et al., 2000).
Like BT, BF placement does not require a titration procedure to determine the seizure threshold. Effective energy dosing is usefully estimated by age and seizure electroencephalogram criteria allowing therapists to avoid the loss of an effective treatment in the first session. This loss occurs when titration methods are required to estimate energy levels for RUL treatment.
Unilateral Electrode Placement
Research interest was galvanized four decades ago with reports that RUL -- both electrodes on one side of the head -- reduced the adverse effects of ECT on memory. Such placement was favored by the first American Psychiatric Association ECT Task Force in 1978. However, clinicians soon found the efficacy of RUL to be considerably less than that of BT. Two new studies found that the inefficacy of RUL was due to its sensitivity to the amount of energy used to elicit the seizure, making it necessary to assess the seizure threshold by a detailed titration method.
In one study, the authors treated four groups of 20 depressed patients each (Sackeim et al., 2000). The two groups treated with RUL at 1.5xST and 2.5xST showed unacceptable remitter rates of 35% and 30%. The group treated with energies 5xST showed a remitter rate of 60%, compared to that of 65% for those treated with BT at 2.5xST. No statistical difference in antidepressant effect was found between high-dose RUL and BT, and the authors concluded that the two treatments were equally effective. However, the samples in this study have been criticized as too small to support the conclusion (Fink et al., in press).
Sackeim et al. also examined the cognitive effects in these patients after two months. They found that in two of nine tests -- the recall of famous events and autobiographical memory -- errors were more persistent in patients treated with BT compared to those treated with RUL.
In a second study, 72 patients with major depression were treated either with titrated RUL at 2.25xST or at a fixed dose set at the 75% energy level (403 mC) of modern brief-pulse ECT devices (McCall et al., 2000). The ST was measured in all patients, permitting the researchers to report the level of energy above ST that was delivered to each. The antidepressant and cognitive effects of RUL at 2.25xST, 3.15xST to 5.04xST, and 8.4xST to 12.6xST were reported. The most effective antidepressant response occurred with energies above 8xST. As energies increased, so did global cognitive performance in an exponentially greater decrement. With the high dose required for efficacy, the advantage in memory effects of RUL over BT placement was lost.
The inefficacy of RUL placement in clinical practice has also been demonstrated in a study of continuation pharmacotherapy after ECT (Sackeim et al., 2001). Depressed patients were mainly treated with RUL placement at the low energy levels recommended in 1993 when the study began. The remission rates for the sample of 290 patients was 55%, an unusually poor rate for conventional ECT. Patients who remitted were asked to participate in a pharmacotherapy continuation treatment program of 24 weeks. The relapse rates were 84% for placebo, 60% for nortriptyline, and 39% for the combination of lithium and nortriptyline. Virtually all patients relapsed within six months.
Summary and Conclusion
These research findings provide guidelines clinicians can use in selecting electrode placement and energy dosing for their depressed patients. Treatment with BT electrode placement is remarkably effective. The remission rates for BT and BF ECT exceed the 50% reduction in HAM-D score that is expected for antidepressant medications marketed with approval of the U.S. Food and Drug Administration. The efficacy of BT ECT for psychotic depressed patients is another encouraging finding, exceeding the success rates reported for antidepressant and antipsychotic drugs when used alone, or in combination, for this condition (Parker et al., 1992).
While the immediate effects on memory may be unpleasant with all electrode placements, those associated with BT are greater than those with BF and RUL. The long-term effects on memory are modest for most patients, even when BT is given at energies that are higher than necessary. Increasing energies in RUL increase efficacy but do not spare cognitive effects. Due to its efficacy and memory-sparing qualities, BF electrode placement is the leading alternative to BT electrode placement in clinical practice today.
.American Psychiatric Association (1978), Electroconvulsive Therapy: Report of the Task Force on Electroconvulsive Therapy of the American Psychiatric Association. Washington, D.C.: American Psychiatric Association.
.Bailine SH, Rifkin A, Kayne E et al. (2000), Comparison of bifrontal and bitemporal ECT for major depression. Am J Psychiatry 157(1):121-123.
.Delva NJ, Brunet D, Hawken ER et al. (2000), Electrical dose and seizure threshold: relations to clinical outcome and cognitive effects in bifrontal, bitemporal, and right unilateral ECT. Journal of ECT 16(4):361-369.
.Fink M, Bailine S, Petrides G (in press), Electrode placement and ECT: a search for the chimera. Arch Gen Psychiatry.
.Lawson JS, Inglis J, Delva NJ et al. (1990), Electrode placement in ECT: cognitive effects. Psychol Med 20(2):335-344.
.Letemendia FJ, Delva NJ, Rodenburg M et al. (1993), Therapeutic advantage of bifrontal electrode placement in ECT. Psychol Med 23(2):349-360.
.McCall WV, Reboussin DM, Weiner RD, Sackeim HA (2000), Titrated moderately suprathreshold vs fixed high-dose right unilateral electroconvulsive therapy. Acute antidepressant and cognitive effects. Arch Gen Psychiatry 57(5):438-444 [see comment pp445-446].
.Parker G, Roy K, Hadzi-Pavlovic D, Pedic F (1992), Psychotic (delusional) depression: a meta-analysis of physical treatments. J Affect Disord 24(1):17-24.
.Sackeim HA, Haskett RF, Mulsant BH (2001), Continuation pharmacotherapy in the prevention of relapse following electroconvulsive therapy: a randomized controlled trial. JAMA 285(10):1299-1307.
.Sackeim HA, Prudic J, Devanand DP et al. (2000), A prospective, randomized, double-blind comparison of bilateral and right unilateral electroconvulsive therapy at different stimulus intensities. Arch Gen Psychiatry 57:425-434 [see comment pp445-446].