Every time a patient with post-traumatic stress disorder has blood drawn for a routine check-up, something is hiding in the numbers. Triglycerides a shade too high. HDL cholesterol a little low. Bilirubin sitting below where you’d expect it. Individually, each reading is easy to explain away. Taken together, across nearly 24,000 patients, they sketch a portrait of a body quietly reorganising itself around a traumatic experience, sometimes years after the fact.
A new study from researchers at Mass General Brigham and Harvard is the most systematic attempt yet to read that portrait. Working through the health records and genomic data of 23,743 biobank participants, they identified 16 laboratory markers that track consistently with PTSD, spanning at least four organ systems simultaneously.
The approach was, in a sense, the opposite of how biomarker research usually works. Instead of starting with a hypothesis, picking a likely suspect and testing it, the team ran what’s called a laboratory-wide association scan: a dragnet across 241 heritable lab traits, looking for anything that moved in step with PTSD. They did this twice over. Once using polygenic risk scores (genetic predisposition to PTSD), and once using actual clinical diagnoses from electronic health records. The 16 markers that came up significant in both passes, concordant in direction, are the ones that deserve serious attention. Not because a single analysis flagged them, but because two independent lines of evidence, one genomic, one clinical, converged on the same signals.
That convergence matters more than it might first appear. It’s roughly the difference between a witness account and a fingerprint match. Each is useful; together, they’re harder to dismiss.
What the 16 markers reveal is that PTSD, whatever else it does to the brain, is doing something to the liver, the blood, the immune system, and the cardiovascular system at the same time. Albumin and bilirubin, both hepatic markers, show characteristic dips. HDL cholesterol falls while VLDL rises, a lipid profile associated with metabolic risk. Mean platelet volume decreases, suggesting altered platelet behaviour. White blood cell subtypes, including lymphocytes, eosinophils, and monocytes, all shift upwards. Vitamin D is lower than expected. So is potassium.
“This multi-system impact helps us understand why untreated PTSD can have such devastating effects on patients’ overall health,” said lead author Younga Heather Lee, an instructor at Mass General Brigham’s Department of Psychiatry. “It underscores the importance of moving away from treating PTSD in isolation and toward recognizing its effects across the body.”
Establishing that these markers associate with PTSD is one thing. Working out which way the arrow points is considerably harder, and it’s where the study gets genuinely interesting. The researchers used a technique called Mendelian randomisation, which exploits the randomness of genetic inheritance to approximate a natural experiment. The idea is to use genetic variants strongly associated with PTSD risk as proxies for the condition itself, then ask whether those proxies predict changes in the biomarkers, or whether biomarker genetics predict PTSD. The answer, across multiple analytical methods, was consistently one-directional: PTSD liability preceded the lab changes, not the reverse. Higher genetic risk for PTSD was associated with lower albumin, lower bilirubin, worse lipid profiles, and lower mean platelet volume. The lab values, in turn, showed no evidence of causing PTSD. The implication is that these biomarkers are probably downstream consequences of the disorder, not pre-existing vulnerabilities that make someone more susceptible to it.
That distinction has real clinical teeth. Downstream consequences are, in principle, things you can monitor and intervene on.
The study can’t yet tell you how early these changes appear, or whether treating them directly would improve psychiatric outcomes. The biobank data represents median lab values across each patient’s entire record, not a temporal sequence. But the direction is suggestive.
“Importantly, our study suggests that PTSD could lead to widespread physical changes affecting cardiometabolic health, immune health, and hepatic health, pointing to PTSD as an upstream contributor to these adverse biomarker profiles,” Lee said. “Finding scalable, blood-based biomarkers could help inform timely interventions aimed at mitigating chronic disease risk, which could ultimately improve long-term health outcomes among patients living with PTSD.”
There are real limitations to reckon with, and the authors are fairly candid about them. The sample was restricted to people of European ancestry, which constrains generalisability considerably given that trauma exposure, care-seeking patterns, and biological risk factors all vary across populations. The biobank skews older (mean age around 61) and toward people with significant healthcare engagement, probably not representative of the many PTSD patients who never make it into a large health system’s records. And the worry about SSRI side effects, which can themselves alter metabolic markers, hasn’t been fully resolved, though the primary associations held up even after adjusting for antidepressant history.
What’s perhaps most striking about the findings is how ordinary the 16 markers are. These are not exotic proteins requiring specialist assays or experimental technology. They’re the readings that come back from a routine blood panel: a cholesterol count, a liver function test, a full blood count. The information, perhaps, has been sitting in medical records for decades. The question is whether anyone has known what to do with it.
https://doi.org/10.1038/s41380-026-03553-z
Frequently Asked Questions
Could a regular blood test one day help diagnose PTSD?
Not as a standalone diagnostic, at least not yet. The 16 biomarkers identified in this study are associated with PTSD, but they also appear in other conditions, so no single blood panel could confirm a PTSD diagnosis on its own. What they might eventually do is help clinicians flag patients at higher risk for PTSD-related physical complications, or track whether treatment is reducing the disorder’s systemic burden. Validation in larger, more diverse populations is the necessary next step.
Why would PTSD affect the liver?
PTSD triggers sustained dysregulation of the body’s stress response systems, including the hypothalamic-pituitary-adrenal axis, which influences a wide range of physiological processes including liver function. Chronic stress-related hormonal changes can alter how the liver synthesises proteins like albumin and processes compounds like bilirubin. The study’s Mendelian randomisation analysis suggests these hepatic changes are a consequence of PTSD rather than a pre-existing vulnerability, though the precise biological pathway remains to be fully mapped.
Is this evidence that PTSD is a physical illness, not just a mental one?
The distinction between “mental” and “physical” illness is increasingly difficult to sustain, and findings like these are part of why. The study shows PTSD producing measurable, consistent changes across four organ systems simultaneously, affecting blood, liver, cardiovascular markers, and immune cell counts. This doesn’t replace the psychological understanding of the disorder; it extends it, suggesting that effective PTSD care may eventually need to address the body as well as the mind.
What’s stopping these biomarkers from being used in clinical practice right now?
Several things. The current study was conducted in a predominantly older, European-ancestry population within a single health system, so it’s not yet clear how well the findings generalise. There’s also the problem of specificity: the 16 markers shift in PTSD, but they shift in other conditions too, making them unreliable as diagnostic signals on their own. Researchers would need to replicate the findings across diverse cohorts, work out the timing of when the changes appear, and figure out how to distinguish PTSD-related patterns from those caused by other factors, including the medications often used to treat PTSD.
ScienceBlog.com has no paywalls, no sponsored content, and no agenda beyond getting the science right. Every story here is written to inform, not to impress an advertiser or push a point of view.
Good science journalism takes time — reading the papers, checking the claims, finding researchers who can put findings in context. We do that work because we think it matters.
If you find this site useful, consider supporting it with a donation. Even a few dollars a month helps keep the coverage independent and free for everyone.