PMDD After a Partial SSRI Response: Luteal-Only Dosing or a Different Drug Class Entirely
Continuous sertraline helped, partially, at a side-effect cost real enough to nearly end the trial — a case about reading a partial response correctly rather than simply calling it a failure and switching drug classes.
Jasmine O., a 26-year-old woman, works as a paralegal at a firm where the billable-hours culture makes it hard to take a bad week off, which is exactly what the ten days before her period have become for the past two years — rage she describes as unrecognizable to herself, crying at her desk, and a drop in concentration severe enough that she's started scheduling her heaviest casework deliberately around her cycle. A structured symptom diary over two cycles confirmed the pattern meets PMDD criteria: symptoms confined to the luteal phase, resolving within a few days of menses onset, with real functional impairment.
She started continuous sertraline 50mg four months ago and the response has been genuinely partial — luteal-phase symptoms measurably less severe by her own diary, but not gone, alongside daily nausea and a flattened libido that persist through the three weeks a month she doesn't actually need symptom relief for. She came in today close to stopping the medication entirely over the side effects, which would discard a drug that is, by her own tracking, doing real work. PMDD's pharmacology is unusual in exactly the way that matters here: unlike depression, where SSRIs take weeks to act, PMDD's serotonergic response is fast enough that dosing confined to the luteal phase alone has shown comparable efficacy to continuous dosing in randomized trials — a genuinely different pharmacokinetic story than most other SSRI indications, and the reason her prior four months isn't simply evidence sertraline should be abandoned.
Jasmine's symptom diary is unusually precise for a self-reported record — she rates severity on a numeric scale twice daily rather than the looser weekly recall most patients bring in, a habit she picked up from tracking billable hours at work and applied here without being asked to. That precision is part of why the team trusts her account of a partial, not absent, response: the data shows her rage episodes dropping from daily to roughly three times per luteal phase, still present but materially less frequent, which is a harder distinction to make from memory alone than from a log kept this consistently. What the diary establishes is that sertraline is working and that its cost is being paid across the three weeks she does not need it; what it cannot establish, because she has never been off it during a luteal phase, is how much of that benefit would survive luteal-only dosing.
Follow-up, close to stopping the medication
Switch to luteal-phase-only dosing rather than switching drug classes. Sertraline is doing real work — her own diary confirms it — and PMDD's pharmacology is genuinely different from depression in this exact way: symptom response is fast enough that confining the dose to the luteal phase alone has shown comparable efficacy to continuous dosing in randomized trials. That keeps the working mechanism while eliminating three weeks a month of side effects she doesn't need to be carrying.
If her cycles were irregular or unpredictable, I'd be much more cautious about this — luteal-only dosing depends on knowing when the luteal phase actually starts, and her cycles being reliably regular is a real precondition here, not incidental.
My concern with luteal-only dosing generally is adherence — it asks a patient to track her own cycle accurately and restart medication at the right point every month, and that's exactly the kind of regimen that quietly fails outside a research setting. A drospirenone-containing combined OCP is FDA-approved specifically for PMDD, works through cyclic hormonal suppression rather than serotonergic modulation, and removes the timing burden entirely — she just takes it daily, continuously.
I'd concede the timing concern matters less here specifically, given how regular and well-tracked her cycles already are — my objection is about the regimen's fragility in general, not necessarily a strong one for Jasmine's actual pattern.
Read her four months for what it actually shows: partial efficacy and a severe side-effect burden, not treatment failure. That combination points specifically at luteal-only dosing of the same drug, which keeps the demonstrated mechanism and directly removes the burden that's driving her toward stopping altogether. Her cycles are regular and well-tracked, which is exactly the precondition Dr. [Psychiatrist] named for this to work reliably. If it doesn't hold up over the next two or three cycles, the drospirenone OCP is a reasonable next step — but abandoning a drug that's working, for a burden that a dosing-schedule change directly solves, isn't the first move.
Agreed: switch to luteal-phase-only sertraline dosing, with a two-cycle symptom diary continued to confirm the reduced-side-effect regimen holds up, and drospirenone/ethinyl estradiol named directly to Jasmine as the fallback if it doesn't. All three voices converged once the disagreement was reframed around what her four-month trial actually demonstrated, rather than treated as a global pass/fail on the drug.