Major Studies on ECT for Depression: What Have We Learned?

These findings were verified in the multisite Consortium for Research in ECT (CORE) collaborative study that used the same populations with the same inclusion and exclusion criteria, evaluations, and time periods as the CUC study.^2-8 After remission, the patients were randomly assigned to continuation treatment with the same combination of lithium(Drug information on lithium) and nortriptyline(Drug information on nortriptyline) or with ECT. The six-month relapse rates for the two treatments were not statistically different from that of the lithium and nortriptyline combination in the CUC study.²

The benefit exhibited with continuation ECT confirms the experience of clinical practice. While medications are easier to administer and are preferred by both patients and practitioners, the efficacy of continuation ECT supports its use in patients who relapse despite the prescription of medications and in those who may not tolerate medications' adverse effects.

The studies differed in treatment strategies. In the CORE study, seizures were induced with bitemporal electrode placement (BT) at 50% above a measured seizure threshold (ST). In the CUC study, electrode placement was right unilateral (RUL), with dosages 150% above the ST.

The CUC collaboration enrolled 349 patients and the CORE enrolled 531 patients with severe depression. Clinical characteristics were comparable, with mean ages ranging from 55 to 59 years; 70% were female and the mean pretreatment Hamilton Rating Scale for Depression (HAM-D) scores were 34 (± 7). At treatment end point (remission), HAM-D scores were between 5 and 6 (± 3) in both studies. The episode duration at the time of referral was from 24 to 31 weeks in the CUC study and from 45 to 49 weeks in the CORE study.

ECT was effective in both studies. Remission—defined as greater than 60% reduction in HAM-D scores and final scores less than 10—was 87% in CORE and 55% in CUC completers. These rates compared favorably with the 30% remission rate in patients taking citalopram(Drug information on citalopram) (Celexa), and the 18%, 21%, and 25% reduction rates in patients tak- ing bupropion (Wellbutrin), sertraline(Drug information on sertraline) (Zoloft), and venlafaxine (Effexor), respectively, in the STAR*D study of patients with nonpsychotic major depression that was diagnosed using DSM-IV criteria.⁹

The difference in remission rates between the CUC and CORE studies is best ascribed to the technical differences in electrode placement and energy dosing. When the CUC study was designed, RUL electrode placement with energy dosing set at 150% above the calibrated ST was considered effective. These settings were used as the primary treatment in more than 90% of CUC study patients. Recent studies found that stimulating energies must be increased considerably in RUL placements, to at least 500% above the calibrated ST, to match the efficacy of BT electrode placement.^10,11 The difference in efficacy is also noted in the number of treatments needed to achieve remission. In the CUC study, the patients received a mean number of 10.2 to 10.8 treatments; of these, 6.6 to 7.7 were RUL treatments. In the CORE study, the mean number of treatments to remission was 7.2 to 7.5—a significant savings in time and cost of an average of three treatments.

In the data analyses, the populations were stratified by the presence of psychosis.² Slightly more than one third of the patients were considered to have psychosis as well as depression. In the CORE study, remissions appeared earlier and were more robust—95% of the patients who had psychosis and depression remitted compared with 83% of patients who had nonpsychotic depression (Figure 1).

In the literature on psychotic depression, the efficacy of antidepressants alone is estimated at 30%, antipsychotics alone at 50%, and the combination of two agents at 70%.¹² The response to ECT is sufficiently greater than that of medications for clinicians to recommend its use as a primary treatment in preference to repeated trials with medications. Such primary use ensures early relief of mood, thought, vegetative signs, and suicide risk.

Suicide is common in persons with depressive mood disorders. The risk is reduced with lithium treatment and with ECT.¹² In the CORE study, 29.5% of the patients expressed suicidal thoughts or reported suicidal acts at baseline.⁴ The HAM-D scores for suicidal intent were reduced to zero in 38% of the patients after one week of treatment, in 61% after two weeks, and in 81% at the end of the course (Figure 2). These findings are supported by comparable CUC data.¹³

The failure of a depressive illness to respond to antidepressant treatments estimated to have been prescribed at adequate dosages for adequate treatment periods has been labeled "treatment resistance." To define treatment adequacy, scientists at Columbia University developed an antidepressant treatment history form (ATHF).¹⁴

In patients treated with ECT, they concluded that those whose depressive illness had failed to respond to adequate pre-ECT pharmacotherapy were substantially less likely to respond to ECT than patients who had not received adequate therapy.¹⁵ In the CORE study, the adequacy of prior treatment bore no relation to treatment efficacy, a finding that is confirmed in other studies.⁸

Using ATHF evaluations, only 2 of 52 patients with psychosis and depression (4%) in the CUC study and only 5 of 106 patients (5%) in the CORE study had received adequate trials of antidepressant and antipsychotic medications before referral for ECT.^6,16 It is not clear whether the clinicians in these leading academic medical centers failed to identify the psychotic form of the depressive illness or had failed to apply adequate treatment algorithms. The identification of psychosis in patients with depression is difficult and is essential in ensuring adequate treatment.