Background
Cerebral white matter changes are common in older adults and are often seen on brain MRI as areas of signal abnormality. Two patterns are especially relevant in aging and cerebrovascular disease: white matter hyperintensities, which are usually more clearly defined, and diffusely abnormal white matter, often described as “dirty-appearing” white matter. Diffusely abnormal white matter, or DAWM, is a subtler MRI finding that may reflect early or mild tissue injury in the brain’s white matter.
White matter contains the nerve fibers that help different parts of the brain communicate. When this tissue is affected by chronic small vessel disease, it can become less efficient, and over time this may contribute to slower thinking, problems with memory, and eventually dementia. However, not every white matter abnormality has the same clinical meaning. The study summarized here asked an important question: in community-dwelling older adults who had relatively limited baseline small vessel disease, is DAWM linked to present-day cognition, future cognitive decline, or later dementia risk?
Why this question matters
Brain MRI often reveals white matter abnormalities long before obvious neurologic symptoms appear. Clinicians already know that larger volumes of white matter hyperintensities are associated with poorer cognitive performance and a higher risk of dementia. DAWM is more subtle and less well studied. If DAWM were a meaningful early marker of cognitive risk, it could help identify people who need closer follow-up or aggressive management of vascular risk factors.
At the same time, it is also possible that DAWM represents an MRI appearance that looks abnormal but does not carry the same prognostic weight as classic white matter hyperintensities. Clarifying this distinction is useful for neurologists, radiologists, and primary care clinicians who interpret MRI findings in older patients.
Study design
This research used data from the Age-Gene/Environment Susceptibility-Reykjavik longitudinal cohort study, a prospective population-based study of older adults living in the community. Participants underwent 1.5T brain MRI scans, including fluid-attenuated inversion recovery, or FLAIR, sequences. FLAIR imaging is particularly useful for detecting white matter abnormalities.
DAWM was visually rated at baseline as a percentage of lobar white matter volume in each brain lobe. The ratings were grouped into four categories: 0%, 0% to 10%, 10% to 25%, and greater than 25% DAWM. The investigators also measured white matter hyperintensity volume at baseline.
The main outcomes were:
1. Cognitive performance at baseline, including memory, executive function, and processing speed.
2. Change in those cognitive domains over about 5.2 years.
3. Dementia status at long-term follow-up, after about 10.3 years.
Because age, sex, education, and vascular risk factors can affect both MRI findings and cognitive outcomes, the statistical models were adjusted for these factors. This type of adjustment helps isolate the association between brain imaging findings and clinical outcomes.
Who was studied
The original cohort included 4,163 participants. For the analyses focusing on relatively limited small vessel disease burden, 2,081 participants were selected based on a median split of baseline white matter hyperintensity volume on FLAIR MRI. The mean age was 74.6 years, and 61% were women.
This is an important detail. The study did not focus on a highly selected group with advanced brain disease. Instead, it examined older adults living in the community, many of whom had only modest white matter changes at the start. That makes the results especially relevant to real-world clinical practice, where MRI findings are often mild or borderline.
Main findings
The key result was clear: baseline DAWM was not associated with cognitive performance at baseline.
Specifically, higher DAWM ratings were not linked to worse memory, executive function, or processing speed scores at the start of the study. The reported associations were small and statistically nonsignificant.
DAWM also was not associated with cognitive decline over the next five years. In other words, participants with more DAWM at baseline did not show a greater drop in memory, executive function, or processing speed than those with less DAWM.
Finally, DAWM was not linked to a higher long-term risk of dementia over about 10 years of follow-up.
By contrast, baseline white matter hyperintensity volume did show the expected associations. Higher WMH volume was associated with worse executive function and slower processing speed at baseline. It was also associated with decline in processing speed over five years and with a higher risk of dementia over 10 years.
This contrast between DAWM and WMH is the most clinically important message from the study.
What the numbers suggest
The reported effect estimates support the idea that DAWM did not carry measurable prognostic value in this cohort. For example, the hazard ratio for dementia associated with DAWM was 0.93, with a confidence interval crossing 1.0, indicating no significant increase in risk.
In contrast, white matter hyperintensity volume had a hazard ratio of 1.35 for dementia, meaning higher WMH burden was associated with a materially increased long-term dementia risk.
This kind of result matters because it shows that not all abnormal white matter findings on MRI should be interpreted the same way. A subtle diffuse appearance may be less worrisome than a larger burden of conventional WMH, especially when the overall burden of small vessel disease is limited.
Possible explanations
Why might DAWM not predict cognitive decline or dementia while WMH does?
One possibility is that DAWM reflects a milder or less specific form of white matter change. It may represent tissue that looks slightly abnormal on MRI but has not yet reached the level of structural damage that disrupts cognition.
Another possibility is that DAWM is more closely related to early or diffuse microstructural changes that are difficult to capture with conventional visual MRI scoring, while WMH volume better reflects established injury. In that case, DAWM might still matter biologically, but its clinical effect could be too small to detect in this study.
It is also possible that DAWM is linked to progression of cerebral small vessel disease, as other studies have suggested, without directly translating into cognitive symptoms unless it progresses further or coexists with a heavier burden of WMH, infarcts, or neurodegeneration.
Clinical implications
For everyday practice, this study suggests a nuanced interpretation of MRI reports.
If an older adult has mild DAWM on MRI but little or no conventional WMH burden, this finding alone may not indicate a clearly increased short- or long-term risk of cognitive decline or dementia. That does not mean the MRI is meaningless, but it does mean clinicians should avoid overinterpreting the term “dirty-appearing white matter” as a strong predictor of dementia.
At the same time, the presence of WMH remains clinically important. White matter hyperintensity burden continues to be a useful marker of brain health and vascular injury, and it should prompt attention to modifiable risk factors such as hypertension, diabetes, smoking, obesity, and physical inactivity.
The findings also support a broader principle in geriatric neurology: imaging abnormalities should be interpreted in context. Age, vascular health, symptoms, cognitive testing, and overall MRI burden all matter. A single MRI descriptor should not be used in isolation to predict a patient’s future.
Strengths of the study
This study had several strengths. It used a large, well-characterized community cohort with long follow-up. It evaluated both cross-sectional cognition and longitudinal outcomes, which provides a fuller view of clinical relevance. It also compared DAWM directly with WMH, making the findings easier to interpret.
Another strength is that the participants were community-dwelling older adults rather than a hospital-based sample. This improves generalizability to many older people who undergo MRI for nonspecific symptoms, screening, or research purposes.
Limitations to keep in mind
As with any observational study, this one cannot prove cause and effect. It shows associations, not direct causation.
The DAWM ratings were based on visual assessment of MRI scans, which is practical and clinically realistic but may be less precise than advanced quantitative imaging methods. Some subtle abnormalities may have been missed or grouped together.
The study also focused on older adults with limited baseline WMH burden. The results may not apply to people with more severe cerebral small vessel disease, stroke, or other neurologic disorders.
Finally, cognition and dementia are influenced by many factors beyond MRI-visible white matter changes, including Alzheimer-type pathology, genetics, mood, sensory impairment, and social or educational factors. Even a well-adjusted study cannot account for every contributor.
Bottom line
In this large community-based cohort of older adults, cerebral “dirty-appearing” white matter was not associated with baseline cognitive performance, cognitive decline over five years, or dementia risk over 10 years. Conventional white matter hyperintensities, however, were associated with worse cognition and higher dementia risk.
The main takeaway is that DAWM appears to be a subtle MRI finding with uncertain clinical significance, at least in older adults with limited baseline small vessel disease burden. It may reflect early white matter change, but on its own it does not seem to predict cognitive decline or dementia in the way that larger white matter hyperintensity burden does.
For clinicians and patients, this is reassuring: not every abnormal-appearing white matter signal on MRI carries the same prognostic weight. For researchers, the study highlights the need for better imaging methods and longer follow-up to determine whether DAWM is simply a mild radiologic marker or an early step in the pathway toward clinically important small vessel brain injury.
Citation
Eiling I, Sigurdsson S, Kuhn-Keller JA, Mooijaart SP, Launer LJ, Van Osch MJP, Gudnason V, de Bresser J. Cerebral “Dirty-Appearing” White Matter in Relation to Cognitive Decline and Dementia Risk in Community-Dwelling Older Adults. Neurology. 2026;106(11):e218036. PMID: 42114038.
