Swedish Registry Study Links Lower Birthweight to Early Adult Stroke

A nationwide Swedish registry analysis found an association between lower birthweight and stroke in early adulthood. The result is notable because the cohort was large and included both women and men, but its meaning is narrower than a claim that cardiovascular fate is fixed before birth.

Researchers examined 768,931 people born in Sweden from 1973 through 1982. By the end of follow-up in 2022, the registries contained 2,252 first stroke events at a mean age of 36. Birthweight below the cohort median of 3.5 kilograms was associated with a 21% higher relative rate of stroke than birthweight above that line after the reported statistical adjustments.

The work was presented at the European Congress on Obesity in May 2026. When the findings were announced in April, they were conference results rather than a peer-reviewed paper. That status does not make the data unimportant, but it limits how far the conclusions can travel into clinical practice.

Nearly 769,000 Records Produced 2,252 Stroke Events

The population-based study included 420,173 men and 348,758 women. Birthweight and gestational-age data came from Sweden's Medical Birth Register, while stroke events were identified through the National Patient Register and Cause of Death Register. Participants were followed through December 31, 2022.

Of the 2,252 first strokes, 1,624 were ischemic strokes, caused by an interruption of blood flow, and 588 were intracerebral hemorrhages, or bleeding within the brain. Forty events were not assigned to either category. The mean age was 37 for ischemic stroke and 33 for intracerebral hemorrhage.

Those event counts represent about 0.29% of the full cohort during the registry follow-up. That crude proportion is not an individual's predicted risk, because follow-up time and personal characteristics matter. It does show why relative and absolute measures should be kept together: a 21% relative difference can be scientifically relevant while stroke remains uncommon in this young population.

The reported association was similar in direction for women and men. Below-median birthweight was associated with an 18% higher relative rate for women and a 23% higher rate for men compared with people above the median. For the cohort as a whole, the reported increase was 21% for all strokes and ischemic stroke, and 27% for intracerebral hemorrhage.

Below the Median Is Not the Clinical Low-Birthweight Cutoff

The terminology needs care. The Swedish comparison divided participants at the cohort median of 3.5 kg. The World Health Organization defines low birthweight as less than 2.5 kg, regardless of gestational age. A person below 3.5 kg in this analysis was therefore not necessarily a low-birthweight infant under the standard clinical definition.

This distinction changes how the result should be described. The study found a gradient associated with lower birthweight and compared two broad halves of the cohort. It did not report that every infant below a recognized medical threshold would have the same adult risk.

Gestational age was not associated with stroke in the reported analysis, and the birthweight association remained after adjustment for gestational age and body mass index in young adulthood. The models also adjusted for sex, birth year, parents' country of birth and age when adult BMI was measured.

Statistical Adjustment Does Not Establish a Cause

An association that remains after adjustment is sometimes described as independent of the included variables. It does not mean that birthweight acts alone, that later health has no effect or that the statistical model removed every possible source of confounding.

Registry studies can identify patterns across large populations, but they cannot by themselves show the biological pathway behind a pattern. The conference report did not measure fetal nutrition, arterial remodeling, organ-cell counts or lifelong epigenetic changes. It therefore cannot support claims that malnourishment permanently altered each participant's blood vessels or that a vascular defect established in the womb caused the later strokes.

The comparison also does not test a screening program. No participants were assigned to have birthweight added to an adult risk calculator, and the study did not report whether doing so improved treatment decisions or prevented strokes. The authors said birthweight may be considered in adult risk assessment; that is a proposal for further evaluation, not a guideline mandate.

A 2021 peer-reviewed Swedish cohort study involving 35,659 men also found an inverse association between birthweight and adult stroke after accounting for young-adult BMI. The newer, much larger analysis extends that line of research to a later birth cohort and women. Publication of the full methods and estimates will still be important for evaluating missing data, model choices and the precision of subgroup results.

The Finding Does Not Erase Adult Prevention

Birthweight is historical and cannot be changed in adulthood. Many established stroke risks can be managed. The US Centers for Disease Control and Prevention identifies high blood pressure as a leading cause of stroke and also lists smoking, diabetes, heart disease, high cholesterol, physical inactivity, obesity and excess alcohol among important factors.

The Swedish result does not show that maintaining a healthy weight, controlling blood pressure or avoiding tobacco is ineffective for people born at a lower weight. It only reports that the birthweight association remained after adjustment for BMI measured in young adulthood. BMI is one measure at one period; it is not a substitute for the full set of cardiovascular risks over a lifetime.

The defensible conclusion is neither dismissal nor destiny. A large registry signal can justify more research into whether early-life information improves risk prediction. But an unpublished conference association cannot declare existing screening tools obsolete, require governments to add birthweight to every adult record or label later stroke as the unavoidable consequence of prenatal conditions.

Risk markers are useful when their size, uncertainty and clinical readiness are stated plainly. Here, the data support an association across a Swedish population with a low absolute event count in early adulthood. They do not support telling an individual that the circumstances of birth sealed the outcome of adult life.