Thursday, August 6, 2026

Opparounds

🔬Psychiatric Research Article

Low-Dose Buprenorphine Following Ketamine Treatment for Suicidal Ideation in Major Depressive Disorder: A Randomized, Double-Blind, Placebo-Controlled Trial

Tucciarone JM, Bandeira ID, Blasey C, Kratter IH, Ehrie J, Keller J, Rodriguez CI, Heifets BD, Schatzberg AF, et al. · American Journal of Psychiatry · June 2026

Stanford investigators gave 50 outpatients with major depression and active suicidal ideation (Scale for Suicide Ideation score of 6 or more) a single open-label ketamine infusion of 0.5 mg/kg over 40 minutes, then 48 hours later randomized them double-blind to four weeks of sublingual buprenorphine 0.2 to 0.8 mg/day or matched placebo. The design tests a specific mechanistic hypothesis: that ketamine's antisuicidal effect runs partly through mu-opioid receptor signaling, and that a partial agonist can hold that signal open after the infusion wears off.

Among the 45 participants who completed at least one week of follow-on treatment, SSI scores fell further with buprenorphine (mean change -11.6) than with placebo (-6.3), a Glass delta of 0.76 with a wide confidence interval of 0.11 to 1.39, and the time-by-treatment interaction was significant (p<0.001). Depression scores did not separate between the groups, so the signal was on suicidal ideation specifically rather than on mood. No serious treatment-related adverse events occurred over the four weeks.

💡 Why it matters

The clinical problem with ketamine for acute suicidality has always been the size of the window, and this is the first randomized evidence that a drug can hold it open — pointing at the mu-opioid system rather than glutamate. It is 50 patients at a single site, prescribing an opioid to a suicidal population, so it is a mechanism worth following, not yet a protocol to copy.

Read the paper → doi:10.1176/appi.ajp.20250840

📖Psychotherapy Research Article

Dialectical Behavior Therapy vs Schema Therapy for Patients With Borderline Personality Disorder: The BOOTS Multicenter Randomized Clinical Trial

Wibbelink CJM, Kamphuis JH, Sinnaeve R, Grasman RPPP, Alberts J, Arntz A, et al. · JAMA Psychiatry · July 2026

BOOTS randomized 204 Dutch outpatients with borderline personality disorder (mean age 32, 84% female) across nine centers to two years of dialectical behavior therapy or schema therapy, each delivered in a combined individual-plus-group format, with blinded assessors and follow-up out to one year after treatment ended. The primary outcome was the rate of change in BPD severity on the BPDSI-5 across the full three years from baseline.

There was no significant difference between the two treatments (P=.27; estimated mean difference at 36 months -1.09, Cohen d=0.15, 95% CI -0.17 to 0.47), and both produced large improvements in severity, functioning, quality of life, well-being, and sleep. No secondary outcome favored either arm. Dropout was substantial and similar in both — 52% by two years in DBT and 46% in schema therapy. The trial was powered for superiority, so the correct reading is an absence of demonstrated difference rather than proven equivalence.

💡 Why it matters

Two structured, well-supervised, two-year treatments built on frankly incompatible theories of what is wrong with these patients produced the same outcome, which moves the clinical question away from which model and toward which patient, which therapist, and what keeps someone in the room. When a patient or a program asks you to pick, the defensible answer is that fit and retention matter more than brand — and that roughly half had left by year two is the number actually worth attacking.

Read the paper → doi:10.1001/jamapsychiatry.2026.0418

💊Psychiatric Fact

The antibiotic on the med list is an MAOI — and the reflex to stop the SSRI is usually the wrong move

Linezolid is an oxazolidinone antibiotic and also a reversible, nonselective MAO inhibitor, which is why the consult arrives phrased as "can we give this to someone on sertraline?" The pharmacology is real, but two decades of case reports badly oversold the risk: retrospective cohorts of co-exposed inpatients consistently find clinically apparent serotonin toxicity in well under 1 percent, and several large series using Hunter criteria find essentially none. Citalopram, escitalopram and methadone sit at the higher end of what little signal exists.

The reason this matters for your recommendation is timing. Stopping the serotonergic agent does not make the interaction go away on the timescale of an infection — you need roughly five half-lives, and with fluoxetine that is about five weeks because of norfluoxetine. There is no acute washout available, so discontinuing usually buys no pharmacologic safety while guaranteeing a destabilized patient who has to be restarted later. When linezolid is genuinely indicated (VRE, or MRSA where vancomycin has failed), the defensible plan is to continue both with an explicit monitoring instruction in the note: serial neurologic exam, not a level — inducible and lower-extremity-predominant clonus, hyperreflexia, then agitation, hyperthermia and autonomic instability. Apply the same logic, with more caution, to methylene blue, a far more potent MAO-A inhibitor that appears unannounced in vasoplegic shock, parathyroid localization and ifosfamide encephalopathy; that one deserves a conversation with the surgical or ICU team before the dose is hung, not after.

🛋️Psychotherapy Teaching Pearl

Analyze the retreat, not the thing retreated from

Defense analysis in the Paul Gray tradition asks you to work at the surface of the hour rather than beneath it. The material of interest is not the warded-off wish; it is the moment the patient turns away from it, live, in the session — the shift from "I was annoyed with you last week" to a story about the parking garage. Gray called the discipline close process monitoring: you follow the flow of associations and mark the point where the flow changes direction, because that point is where the ego did something, and what the ego does is the one part of this the patient can observe alongside you.

The technical claim is that interpreting the content behind the defense — the anger, the wish, the envy — hands the patient a conclusion instead of an experience. What comes back is compliance ("yes, I suppose I was angry") or a firmer defense, and either way the patient learns that you are the one who supplies meanings. Marking the shift itself recruits the observing ego as a collaborator and makes the governing anxiety available: not "you were angry with me," but "what made it necessary to move away from that, here, with me?" The order is the technique — anxiety and defense before impulse, surface before depth.

🗒️ Vignette

A resident's long-term patient, a 40-year-old man in twice-weekly work for chronic dysthymia and a lifelong pattern of quiet deference, opens the hour: "I was actually irritated when you ended five minutes early last week — you probably had somewhere to be." Then, with no pause: "Anyway, the parking garage has been a nightmare, they closed the second level."

The pull is to interpret the transference — he is angry at you, this is his father again, he cannot tolerate his own aggression. Plausibly all true, and all premature: offered now it arrives as your insight about a feeling he has just demonstrated he cannot yet hold. The resident instead marks the movement. "Something happened just then. You told me you were irritated with me, and then you were telling me about the garage. Did you notice it?" He goes quiet, then says he did notice — that the moment the words were out he thought he had been unfair, that you surely had a good reason. The self-correction is the defense, and it just happened in the room where both of them could watch it.

Only from there does the resident ask what he imagined would happen if the irritation had been allowed to stand — the anxiety the defense exists to manage. He says, with real surprise, that he pictured the therapist becoming quietly cooler with him for weeks afterward. That fantasy, not the anger, is what has organized thirty years of deference, and it surfaced only because the anger went uninterpreted.