Ketamine for Treatment-Resistant PTSD
Single-dose IV ketamine produced rapid, if transient, PTSD symptom reduction in a placebo-controlled trial. Whether that justifies off-label use in a patient whose most disabling symptoms are dissociative is a genuinely separate question from whether ketamine works.
S.B. is a 33-year-old woman, a hospice social worker, whose PTSD followed a violent home invasion four years ago; she still works full-time, a fact she raises unprompted because she worries the team will assume she is more impaired than she is. She has completed three adequate antidepressant trials — sertraline, venlafaxine, and fluoxetine, each at maximal tolerated doses for at least ten weeks — and a full course of both prolonged exposure and EMDR, with only modest and short-lived benefit from any of them. She has no substance use history and no psychotic symptoms, both directly relevant to what she is asking about today.
What distinguishes her presentation, and complicates the option she has read about, is a prominent dissociative subtype of her PTSD: depersonalization and emotional numbing severe enough that she has described "watching myself from outside my body" during flashbacks, a symptom pattern that predates any medication trial. She is asking about ketamine after reading about the 2014 placebo-controlled crossover trial that found a single IV infusion produced rapid, clinically meaningful PTSD symptom reduction compared to an active midazolam control, measured at twenty-four hours. That trial did not follow responders long enough to establish how long the benefit lasts; the impression that it is transient comes from the wider literature and from a later repeated-dose trial, in which responders to a two-week infusion course relapsed at a median of about four weeks. The tension is not about whether ketamine can work — the trial data on rapid symptom reduction is real — it is that ketamine itself reliably produces dissociative effects during infusion, and giving a drug whose acute mechanism includes dissociation to a patient whose most disabling baseline symptom is already dissociation is not a contradiction the original trial was designed to address, since dissociative-subtype PTSD was not specifically isolated in that population.
At the follow-up visit
The 2014 trial data is real and worth taking seriously — a single infusion producing rapid, placebo-controlled symptom reduction in a genuinely treatment-resistant population is not a small finding, even though it tells us almost nothing about durability. But that trial was not designed around dissociative-subtype PTSD specifically, and giving a drug that reliably induces dissociation during infusion to a patient whose core symptom is already dissociation is a real, mechanism-level concern that population-level results don't settle for her.
I take the concern seriously, but I don't think it resolves cleanly against her either way. Some clinicians treating dissociative-subtype PTSD have reported the induced dissociation is experienced as qualitatively different from her spontaneous flashback dissociation — controlled, time-limited, in a monitored setting — rather than a re-triggering of the same symptom. That is a real distinction, but it is also not something either of us can promise her holds true in her specific case.
Given three failed adequate trials and two failed psychotherapy courses, I don't think withholding a genuinely evidence-backed option from her on a theoretical concern we can't confirm is obviously the safer choice either.
If we proceed, it should be a single monitored infusion first, not a course committed to in advance, with explicit real-time tracking of whether the induced dissociation reads to her as distinct from her baseline symptom or as a direct trigger of it. That single data point, from her own experience, is worth more than either of our predictions about how it will go.
A single monitored IV ketamine infusion scheduled, explicitly framed as a trial rather than the start of a committed course, with S.B. informed in detail about the expected acute dissociative effects beforehand.
Left honestly unresolved: whether her dissociative-subtype presentation will make the infusion feel like a re-triggering of her worst symptom or a distinct, tolerable experience. Both the team and S.B. agreed the single infusion itself is the only way to actually find out, rather than reasoning it out from a trial that was never designed to answer this specific question.