PMDD: Luteal-Phase-Only vs. Continuous SSRI Dosing
SSRIs act fast enough in PMDD to be dosed only in the luteal phase — a real option for a patient with predictable cycles, and a much harder one for a patient whose cycles won't reliably tell her when the luteal phase begins.
R.J., a 26-year-old marketing analyst, has tracked two full menstrual cycles on a DRSP diary at her gynecologist's request, confirming a pattern that has been disrupting her work and relationships for about three years: irritability, tearfulness, and marked anxiety that arrive reliably five to seven days before her period and resolve within a day or two of onset, with no significant symptoms at any other point in her cycle. Her cycles have run a consistent 28 days, give or take a day, for as long as she has tracked them, and she has no other chronic illness and no history of depression outside this cyclical pattern. Her mother had a similar premenstrual pattern that was never formally diagnosed or treated, something R.J. only learned after describing her own symptoms at a family dinner, and which she says made her finally take the pattern seriously enough to seek a diagnosis rather than continue attributing it to ordinary stress.
She works a job with a predictable schedule, has never missed a dose of any medication she's been prescribed before, and has specifically asked whether she needs to take anything every day or only when symptoms are expected — she'd prefer to avoid a daily medication if a more targeted option is genuinely equivalent. SSRIs are unusual among antidepressants in that their effect in PMDD can appear within days rather than the several weeks typically needed in major depression, a distinct pharmacology that is what makes luteal-phase-limited dosing a real, evidence-supported option rather than a compromise — provided the luteal phase itself can be predicted reliably enough to start the drug on time.
At the treatment-planning visit
Her cycle regularity is exactly what makes luteal-phase-only dosing pharmacologically sound rather than a compromise — sertraline starting on approximately cycle day 14 through the onset of menses covers her symptomatic window with real margin, and PMDD's rapid-onset SSRI response, thought to work partly through rapid enhancement of allopregnanolone, a progesterone metabolite that modulates GABA-A, rather than requiring the weeks of monoaminergic adaptation depression treatment needs, is what makes starting mid-cycle actually effective rather than too late.
Given how consistent her cycles have been and her clean adherence history, intermittent dosing is a reasonable match for both the pharmacology and her stated preference — I'd build in one safeguard, a simple ovulation-predictor or basal-temperature check as a backup cue, so a single unusually long cycle doesn't leave her starting a few days later than her symptoms actually do.
Sertraline 50mg was started on a luteal-phase-only schedule, timed to approximately cycle day 14 through menses onset, with a basal-temperature or ovulation-predictor check added as a backup cue against an occasional longer cycle.
The team's confidence in this approach rests specifically on her cycle's demonstrated regularity — a fact that becomes the entire subject of Patient B's case below, where the same drug and the same diagnosis lead to a different dosing strategy.
M.A., a 44-year-old high-school administrator in early perimenopause, has the same DRSP-confirmed diagnosis as Patient A — irritability, tearfulness, and anxiety that a two-cycle diary places in the days before menses — but her cycles have ranged from 21 to 45 days over the past year, with no consistent pattern her gynecologist or she can use to predict when the next luteal phase will actually begin. She has no other chronic illness, and this is her first time seeking treatment for a premenstrual mood pattern that, by her account, has worsened noticeably as her cycles have become less predictable over the past eighteen months. She manages a school office with a demanding public-facing schedule, and has described specific incidents — snapping at a parent during a routine call, canceling a staff meeting she'd normally handle without difficulty — that she now recognizes, after the diary, as falling in the same unpredictable premenstrual window each time.
The diagnosis itself isn't in question — her diary shows the same luteal-restricted symptom clustering as Patient A's, just measured against a far less predictable cycle. The problem is purely one of timing: luteal-phase-only dosing depends on knowing when the luteal phase starts closely enough to begin the drug in time, and a cycle that swings between three weeks and six weeks doesn't give her, or the team, a reliable cue to dose against. Starting too early wastes days of unnecessary medication; starting too late means missing the window the drug is meant to cover entirely, which is functionally the same as not treating her at all that month.
At the treatment-planning visit
I'd default straight to continuous dosing for her rather than trying to make intermittent dosing work around an unpredictable cycle — a missed or mistimed luteal window isn't a minor inefficiency, it's a month where her PMDD goes essentially untreated, and that's a worse outcome than accepting daily medication.
I agree with the direction, and it's worth being specific about the tradeoff rather than treating it as free — continuous dosing means she's exposed to the SSRI's side-effect profile every day of the month rather than the roughly two weeks — cycle day 14 through menses — that Patient A is on it, for a diagnosis whose symptoms only appear in that same restricted window.
Reproductive endocrinology consultation to characterize how consistently perimenopausal her cycles actually are might eventually clarify whether some luteal-timing strategy becomes feasible again later, but that's a future reassessment, not something to build today's plan around.
Sertraline 50mg was started as a continuous daily dose rather than timed to the cycle, with a referral to reproductive endocrinology to characterize her perimenopausal pattern more precisely.
The team was explicit that this was the same diagnosis and the same first-line drug as Patient A, arriving at a different regimen for one specific, identifiable reason — nothing about M.A.'s PMDD is more severe or harder to treat pharmacologically; her cycle simply doesn't give the team anything reliable to time a smaller dose against.