Patients with severe melancholic or psychotic depression are more likely to respond to ECT, and preliminary evidence indicates they’re also protected against ECT-induced cognitive impairment, Linda van Diermen, MD, PhD, reported at the virtual congress of the European College of Neuropsychopharmacology.
Dr. Linda van Dierman
Over the decades many small, underpowered studies have looked at possible predictors of ECT response and remission, with no consensus being reached. In an effort to bring a measure of clarity, Dr. van Diermen and her coinvestigators performed a meta-analysis of 34 published studies in accord with the PRISMA-P (Preferred Reporting Items for Systematic Review and Meta-analysis Protocols) guidelines and published their findings in the British Journal of Psychiatry. They scrutinized three potential predictors of response: the presence of psychotic features, melancholic depression with psychomotor symptoms, and older age.
Psychotic depression was associated with a 1.7-fold increased likelihood of response to ECT and a 1.5-fold increased odds of remission, compared with that of ECT-treated patients without psychotic depression. Older age was also a statistically significant predictor of response. However, the findings on melancholic depression were inconclusive, with only five studies with inconsistent results being available, said Dr. van Diermen, a psychiatrist at the University of Antwerp (Belgium).
She was quick to point out that, although psychotic depression and older age were statistically significant predictors of heightened likelihood of ECT response, they are of only limited clinical significance in treatment decision-making. The ECT response rate was 79% in patients with psychotic depression but still quite good at 71% in those without psychotic depression. Moreover, the average age of remitters was 59.7 years, compared with 55.4 years in nonresponders, a difference too small to be useful in guiding clinical treatment decisions.
“Although we did a meta-analysis in more than 3,200 patients that confirmed the superior effects of ECT in older patients and we recommended it at that time as one of the elements to guide decision-making when you consider ECT, our present, more detailed look at the interdependence of the predictors leads us to reconsider this statement. We now venture that age has been given too much weight in the past decades.”
A closer look at ECT response predictors
The studies included in the meta-analysis assessed psychotic depression and melancholic features as ECT response predictors in the typical binary way employed in clinical practice: yes/no, either present or absent. Dr. van Diermer hypothesized that a more in-depth assessment of the severity of those factors would boost their predictive power.
She found that this was indeed the case for melancholic depression as evaluated by three tools for measuring psychomotor symptoms, a core feature of this form of depression. She and her coinvestigators assessed psychomotor functioning in 65 adults with major depressive disorder before, during, and after ECT using the clinician-rated CORE scale, which measures psychomotor retardation, agitation, and noninteractiveness. In addition, the investigators had the subjects wear an accelerometer and complete a timed fine-motor drawing test.
The 41 patients with melancholic depression with psychomotor symptoms as defined by a CORE score of 8 or more were 4.9-fold more likely to reach an ECT response than were those with nonmelancholic depression. A lower baseline daytime activity level as assessed by accelerometer was also a significant predictor of increased likelihood of response, as were slower times on the drawing test.
In contrast, the investigators found that more detailed assessment of psychotic depression using the validated Psychotic Depression Assessment Scale (PDAS) was predictive of the likelihood of ECT response, but not any more so than the simple presence or absence of psychotic symptoms (J ECT. 2019 Dec;35[4]:238-44).
“In our sample, better measurement of psychotic symptoms did not improve prediction, but better measurement of psychomotor symptoms did seem to be valuable,” according to the psychiatrist.