Science2 publishers2 min readPublished
Age-related gray-matter loss pauses in the 120 women who crossed into menopause mid-study
A two-scan study of 1,095 female brains in Nature Communications finds the ordinary adult decline in cortical gray matter flattening through late perimenopause and resuming afterward. Its first author says the implications are hard to state.
The Scientist · Science desk
What happened
- A study published Sept. 8 in Nature Communications compared MRI scans taken at two time points from 1,095 brains spanning puberty, pregnancy and the transition into menopause.
- The gradual adult decline in gray matter leveled off during late perimenopause and resumed in postmenopausal women, with the timing of the pause varying between women but showing up across the group.
- The menopause scans came from the UK Biobank, a database covering 500,000 U.K.-based adults, and the puberty and pregnancy scans came from datasets compiled at Leiden University in the Netherlands.
- Girls entering puberty and women who were pregnant lost gray matter, matching what earlier studies of those two transitions had found.
- The menopause result does not entirely agree with previous research that found distinct reductions in hippocampal gray matter volume during the transition.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- constraint The study measured structure and not cognition, so it gives a clinician no way either to attribute a midlife memory complaint to volume loss or to rule that out; its first author says the link to cognitive function cannot yet be stated.
- contradiction A stable cortical total and a shrinking hippocampus can both be true at once, so whether the transition reads as benign depends on which measurement a trial or a clinic adopts as its endpoint.
- decision Anyone treating the menopausal transition as a window of neurodegeneration now needs a measure other than overall cortical gray matter volume to carry that claim.
- capability With one two-scan comparison covering puberty, pregnancy and menopause, later work can ask whether the three hormonal transitions differ in mechanism inside a single dataset.
The plateau rests on 120 women. Each was scanned twice, about 3.5 years apart, and between the two scans each passed the full year without a period that defines menopause [8][9]. That within-person design is why the finding is worth reading. It compares a woman with her own earlier scan, so it does not have to assume that two women of different ages differ only in menopause status.
The reference groups carry that other assumption, and they are lopsided. About 50 women who had not yet had a last period, mean age 51, sit against more than 670 who had, mean age 55.5 [10]. Their mean ages are 4.5 years apart [3].
Add the three menopause groups together and they account for roughly 840 of the study's 1,095 brains [1]. The pregnancy arm is 110: 40 first pregnancies during the study, 30 second pregnancies and 40 women who had never been pregnant [2][22].
"No one had looked at the changes in the whole brain matter in this menopause transition before," said Sophie van 't Hof, a doctoral student in psychiatry at Amsterdam University Medical Center and the study's first author, at a press conference [13][5].
What the scans do not speak to is thinking. Van 't Hof said it is too soon to say whether any specific structural change is linked to cognitive function in menopause [17]. She has made the same point about pregnancy more bluntly: "But what's really important is that gray matter decline is not equal to cognitive complaints" [25]. The losses at puberty and in pregnancy, she said, look to her team like "neural fine-tuning" [24].
The measure here is total cortical gray matter volume, the wrinkled outer layer that contributes to memory, reasoning and movement [1][12]. A total can hold steady while one structure inside it loses volume.
Menopause stage came from what participants reported about their own periods, and the researchers did not confirm it with hormone or other diagnostic tests [15]. So the line between late perimenopause and postmenopause is drawn from recall.
Magdalena Martínez-García, scientific director of maternal health for the Ann S. Bowers Women's Brain Health Initiative, who was not involved in the work, told Live Science by email that "This study provides the first direct longitudinal comparison of brain structural changes across all three major female hormonal transitions" [19][20]. On what follows from it, van 't Hof said: "The exact implications are very difficult at this stage" [18].
What to watch
- A replication that stages menopause with hormone assays instead of self-report would test whether the pause falls where this study puts it.
- A longitudinal hippocampus-specific analysis paired with cognitive testing would show whether the regional and whole-cortex results can both hold.
- Whether the per-woman timing of the pause lines up with when symptoms start.