FSH Testing in Perimenopause: Why One Test Isn’t Enough

Why One FSH Test Isn’t the Answer in Perimenopause

Written by Carrie Jones, ND, FABNE, MPH, MSCP

 

FSH is having a quiet moment in longevity medicine — though most of us have not noticed yet.

Follicle-stimulating hormone is still treated, in most practices, as a simple yes-or-no test: ordered once, or at most twice, to confirm menopause (or evaluate fertility), then set aside. But a growing body of research is reframing FSH as something more interesting: a longitudinal signal that may track how a woman’s bone, muscle, brain, and metabolism are aging. These are the very tissues we now think of as longevity organs. If that reframe holds, then the way most of us use FSH is leaving a great deal of information on the table.

I want to be upfront that much of this is still emerging science. Some of it is more established; some rests on animal and cell models with human data that are still largely observational. I will flag which is which. But the direction of travel is clear enough to raise a clinical question we rarely ask: what happens when we treat one of the most dynamic hormones of midlife as a single, static number?

 

A Test We Order Once and a Hormone That Never Sits Still

Perimenopausal hormones are, to put it politely, chaos. Estradiol does not glide gently downward the way textbook curves suggest; it spikes and crashes, sometimes within a single week. FSH does the same, out of sync. As the 2025 European Society of Endocrinology guideline puts it plainly: blood tests are not recommended to diagnose perimenopause because serum estradiol levels fluctuate and can be misleading, and a “normal” level does not rule anything out.

Here is why a single draw falls short in the years when it matters most:

• Near-random reliability. The reproducibility of a single FSH measurement in premenopausal women is famously poor — with a reliability coefficient of around 0.09, versus 0.70 once a woman is firmly postmenopausal.
• The societies already agree. The Menopause Society, ACOG, the International Menopause Society, and the STRAW+10 staging system all advise against using a single FSH value as a standalone diagnostic in perimenopause. Where FSH is used to help confirm early or premature menopause, the guidance is to repeat it, typically two draws four to six weeks apart, and for women over 45 with typical symptoms, to skip the test and diagnose based on symptoms alone.
• Two moving numbers, one snapshot. A transiently high estradiol can suppress FSH into the “normal” range, so even drawing both on the same morning can reassure you about a woman whose ovaries are actively winding down. And two draws six weeks apart are still just two frames of a hormone that can swing day to day.

In other words, the problem is not the assay. It is that we are trying to describe a moving target with one or two photographs.

 

Bone, Muscle, and Brain Are Longevity Organs — and FSH May Track Them All

The reason to care about FSH beyond menopause timing is that its receptors turn up well outside the ovary on bone-resorbing cells, fat cells, immune cells, blood-vessel lining, and, in animal models, in the brain. A landmark 2006 Cell paper showed FSH can act directly on bone, independent of estrogen. Since then, the same finding has kept reappearing across very different tissues.

Here is where it shows up — with my honest read on how strong each line of evidence is:

• Bone (well established). In the SWAN cohort, bone density begins declining while FSH is rising but estradiol still looks normal; bone loss tracks the rising FSH more closely than the falling estrogen. The unsettling implication: bone loss may be underway while your patient still has regular periods.
• Muscle (emerging). This is the newest one that pulled me in. The June 2026 Menopause editorial asked, “Is sarcopenia the new marker for postmenopausal women’s health?” — noting that 28% of older women with obesity meet criteria for sarcopenic obesity, which carries higher mortality than either condition alone. I found emerging evidence that higher FSH tracks with lower lean mass even after adjusting for estrogen and testosterone (though in men from that same cohort, no such link was found, suggesting the effect may be specific to postmenopausal physiology).
• Brain (frontier). Women carry roughly twice the risk of Alzheimer’s after menopause compared with men, and estrogen loss alone has never fully explained that gap. A 2025 study of 884 postmenopausal women found FSH, not estradiol, was associated with poorer cognition and higher amyloid burden. In mice, FSH acts on hippocampal neurons, and blocking it reverses amyloid and tau pathology. Compelling, but not guideline-ready.
• Inflammation (emerging). Underneath all of it runs inflammaging. In cell and animal studies, FSH stimulates the same cytokines we chase clinically — TNF-alpha , IL-1beta, IL-6 — through its own receptors on immune cells, independent of estrogen. In a small randomized trial, suppressing gonadotropins (including FSH) lowered inflammatory markers and improved disease activity in rheumatoid arthritis.

No single finding here rewrites practice. But taken together, they make FSH look less like a menopause “yes or no” checkbox and more like a continuous readout of biological aging.

 

The Right Clinical Question

Instead of only asking “Is she in menopause yet?” — a binary a single FSH cannot reliably answer — the more useful question becomes: where is her FSH heading over time, and what might that trajectory be doing to her bone, muscle, and brain along the way? That is a longevity question, and it is not one you can likely answer from a single blood test while she’s in perimenopause.

 

From Snapshot to Trend: What Continuous Monitoring Reveals

We solved a version of this problem once already. We do not manage diabetes with a fasting glucose every three months; we can use continuous data, because the pattern — the peaks, the troughs, the trend — is what guides treatment. The same logic is finally reaching hormone care.

Mira is a quantitative at-home monitor that measures urinary E3G (an estrogen metabolite), LH, PdG (a progesterone metabolite), and FSH daily using fluorescence-based immunoassay detection, the same class of technology used in laboratory immunoassays. It lets us watch the trajectory instead of guessing at it from a snapshot in time.

What continuous data catches that a single draw misses:

• Anovulatory cycles — often missed entirely in standard care
• A rising FSH trend, months before a single draw would flag it
• Erratic, non-cyclical estrogen — the surges and crashes behind many symptoms
• Symptoms mapped to where she actually is in her cycle, or is not

Consider the patient I see most often: 46, still cycling, presenting with insomnia, mood swings, brain fog, and new joint aches. Her FSH comes back at 12 and her estradiol at 89, and she is told she is “not in menopause yet.” The snapshot said fine. Longitudinal data tells a different story — though what that story reveals varies by patient. Here is what it can look like.

 

[FIGURES 1 & 2] The same 46-year-old, two months apart. In the first cycle, FSH (yellow) stays within reference range, rising to 10.9 on cycle day 4, while estrogen (black) coordinates cleanly into an LH surge (turquoise) and ovulation. One month later, FSH is running higher through the early cycle, estrogen is lower, and ovulation still happens — but the hormone pattern has shifted. FSH can look entirely normal in one cycle and be significantly elevated in the next; estrogen can look robust one month and run lower the next, even as ovulation continues. Snapshot blood work — drawn once, in either month — would never have caught either shift. Only watching both did.

 

[FIGURE 3] A 49-year-old whose estrogen (black) sits within an average range for the first 20 days of the cycle, then rises to an abnormally high level and stays there, without the LH surge (turquoise) that would normally accompany it. No ovulation follows. That sustained high estrogen tracks with, and helps explain, a rise in her symptoms, giving her a reason for what she’d been feeling. A one-time blood draw, taken almost anywhere in those first three weeks, would have shown nothing out of the ordinary and left her symptoms unexplained.

 

[FIGURE 4] A 45-year-old with two starkly different pictures depending on when you catch her: early on, low estrogen (black) paired with high FSH (yellow); later, the reverse — estrogen abnormally high, FSH low. Neither hormone ever coordinates enough to trigger ovulation. Test her on a random day and you could just as easily conclude she’s estrogen-deficient as estrogen-excess, and her symptoms would look just as different depending on which one is true that week.

 

Each of these is visible within a cycle or two once you’re looking at daily data instead of a single point. What takes longer, in my experience, is the full picture — it’s usually months, not weeks, before daily data moves from “some irregularity” to a clear late-perimenopausal pattern. That longer arc is exactly why a snapshot is the wrong tool for seeing it.

 

When to Track FSH Longitudinally

For most perimenopausal patients, a single FSH adds little. Certain presentations, though, are exactly where a trend, rather than a snapshot, changes management. Consider longitudinal monitoring when you see:

• Classic symptoms with “normal” or contradictory single-draw labs
• A family history of osteoporosis or dementia, where the intervention window matters
• Early changes in strength, body composition, or recovery
• A patient starting, adjusting, or monitoring hormone therapy
• Anyone whose symptoms and labs simply do not line up

For clinicians who want to build this into practice, Mira offers a clinician partnership program with a dashboard, patient reports, and clinical support; you can create a practitioner account and download the companion FSH clinical guide by meeting with Mira’s clinical team.

I’ll be going much deeper into all of this in the upcoming Functional Medicine IS Longevity™ Masterclass for clinicians, where I’ll walk through real longitudinal FSH data alongside the Mira team and show how to fold it into an existing workup. If you care about where menopause care and longevity medicine intersect, it is a conversation worth joining.

 

 

Back to the Right Question

I do not want to oversell FSH. The science is moving quickly, some of it will be revised, and a good clinician holds emerging evidence loosely. But I also do not want us to keep treating one of the most dynamic signals of midlife as a one-and-done checkbox. Menopause is the biological inflection point for bone, muscle, brain, and heart — and for most patients it goes almost entirely unmonitored.

For most of my career, we watched that inflection point in the rear-view mirror. For the first time, we can watch it as it happens. That is what makes this such an interesting moment to practice, and why the better question is no longer whether a woman is in menopause, but where her hormones are heading, and whether we are paying close enough attention to do something about it in time.

Disclosure: This article was produced in partnership with Mira. The clinical interpretations and opinions are the author’s own, and it is educational rather than a substitute for individualized medical advice.

 

Explore Mira’s clinician partnership program and create your practitioner account here. And download the companion FSH clinical guide for a concise summary of how to identify the right patients, interpret longitudinal hormone data, and fold it into your existing workup.