Thursday, October 29, 2026
Two papers, a psychopharmacology pearl, a psychotherapy pearl and the news.
AI-generated content. This briefing was written by Claude Opus 5.5 and may contain errors. Please verify against the primary source before acting clinically.
| The Paper |
| New research from the psychiatric journals |
Clinical and Cognitive Outcomes Comparing Right Unilateral Ultrabrief Electroconvulsive Therapy Versus Magnetic Seizure Therapy for Bipolar Depression: The CORRECT-BD Trial
Blumberger DM, Vila Rodriguez F, McClintock SM, Thorpe KE, Burhan AM, Foley K, Goodman MS, Gregory EC, Kaster TS, Knyahnytska Y, Tham J, Trevizol AP, Voineskos D, Zrenner B, Daskalakis ZJ. American Journal of Psychiatry. October 2026 (online May 2026).
Magnetic seizure therapy (MST) induces a seizure with a focused magnetic field rather than electrical current, so less current reaches the hippocampus and medial temporal structures thought to drive ECT's memory effects. In this double-blind, randomized, parallel-group pilot trial at Canadian academic centres, 55 adults with bipolar depression were randomized to right unilateral ultrabrief-pulse (RUL-UB) ECT or MST and treated until remission, dropout, or a maximum of 21 sessions.
Among the 45 who received an adequate course, remission on the HAM-D-24 occurred in 6 of 20 (30%) with ECT and 5 of 25 (20%) with MST, and secondary clinical outcomes were similar. Clinically important autobiographical memory loss (a 25% or greater drop on the Autobiographical Memory Test) occurred in 22.2% of ECT patients versus 7.1% with MST, and MST was better tolerated. The study was a pilot and was not powered to show non-inferiority.
|
In practice
When a patient with bipolar depression refuses ECT because of memory fears, MST is a plausible future alternative, but this pilot cannot tell you it works as well, so it belongs in a research referral rather than standard care. The comparator also matters: ECT here was already the memory-sparing right unilateral ultrabrief form, so the memory gap is not against bitemporal ECT. |
Full text doi:10.1176/appi.ajp.20250955
| The Therapy Evidence |
| Research on what actually works in the room |
Using machine learning to predict long-term treatment outcomes in psychotherapy for personality disorders
Van Den Eshof JW, Leferink RG, Van HL, Van Grieken RA, Arntz A, Cremers HR. Psychotherapy Research. September 2026.
This Amsterdam study followed 524 patients treated in a specialized outpatient personality-disorder psychotherapy program and measured personality functioning one year after treatment ended. The authors compared simple regularized regression with more complex tree-based machine-learning models, first using only intake variables and then adding variables measured at the end of treatment.
Intake data alone predicted one-year outcome poorly, with a mean absolute error of 0.40 to 0.44 on a 0 to 4 scale. Adding end-of-treatment variables cut the error to 0.27 to 0.33, and only then did the complex models do slightly better than the simple ones. The variables that carried the prediction were anhedonia, self-esteem, and relational functioning at treatment end.
|
In practice
Intake severity tells you little about who will hold their gains a year after personality-disorder treatment; how the patient looks at the end, especially in pleasure, self-regard, and relationships, tells you much more. Before discharge, look specifically at those three domains, because residual problems there are the strongest signal that booster sessions or a step-down plan are needed. |
Full text doi:10.1080/10503307.2026.2721657
| Past the Boards |
| One advanced clinical point, usually pharmacology |
Olanzapine pamoate's observation period exists because a fraction of the depot can enter a vein
Post-injection delirium/sedation syndrome (PDSS) after long-acting olanzapine pamoate looks like an olanzapine overdose: sedation, confusion, slurred speech, ataxia, and sometimes unconsciousness, without clinically significant drops in vital signs. The probable mechanism is accidental entry of part of the dose into a blood vessel, where the pamoate salt dissolves far faster than it does in muscle. In the registration data, about 45,000 injections given to 2,054 patients, it occurred in 0.07% of injections, or 1.4% of patients, and no clear risk factors were found. That means risk builds with every injection, and a patient who has had many uneventful shots is not protected. Median onset was 25 minutes, but onset ranged up to 3 to 5 hours, which is why the label requires a supervised post-injection observation period. Every patient recovered, within 1.5 to 72 hours, and most continued the drug. There is no antidote, so care is supportive. In practice, aspirate before injecting, use a fresh site each time, and treat any new drowsiness in the observation window as PDSS until proven otherwise.
| In the Room |
| A psychotherapy technique, shown in a vignette |
An enactment you have already joined is better material than one you avoided
Enactments are jointly constructed: the patient's unconscious pressure meets the therapist's own valency, and the therapist acts before they think. Trainees are often taught to avoid them, but the more useful skill is recovering from one, because the enactment puts the patient's object relation into the room where both people can study it. The technical sequence is to notice your own participation first, to name what happened between you without blame and without a confession that turns the hour toward your feelings, and only then to wonder what the scene repeats. Interpreting the patient's part before acknowledging yours usually reads as the therapist blaming the patient for something the therapist did, which replays the original relationship. Owning a piece of it, briefly and concretely, models that a relationship can hold a mistake. That is often the lesson the patient never got.
|
Vignette
A man in his 30s with a critical, unpredictable father speaks so tentatively that the resident has started finishing his sentences. One day the resident finishes one wrongly, and the patient quietly agrees with the wrong version. The resident catches it next session: "Last week I put words in your mouth that weren't yours, and you went along with them. I've been doing that for a while. I wonder whether going along felt safer than correcting me." The patient pauses, then says that correcting his father "never ended well." The resident does not apologize further or explain the habit. Owning the move first made it safe for the patient to look at his own compliance, and the pattern became a shared object rather than an accusation. |
| *** |
| That's the five. Go see your first patient. |