CT-Detected White Matter Disease Signals Higher Stroke/Dementia Risk Than MRI-Only Findings

Navigating Incidental Findings: The Clinical Significance of Covert Cerebrovascular Disease

Covert cerebrovascular disease (CCD), which includes covert brain infarction and white matter disease, is a frequent incidental finding on neuroimaging ordered for unrelated reasons. While often clinically silent at the time of discovery, these brain changes are increasingly recognized as significant precursors to ischemic stroke and dementia [1, 2, 3, 4]. Clinicians face a practical challenge: while magnetic resonance imaging (MRI) is highly sensitive for detecting such pathology, computed tomography (CT) is more commonly used in routine care [4]. Understanding the discrepancies in detection between these modalities, and more importantly, the distinct prognostic weight of findings from each, is essential for accurate risk stratification and patient counseling [5, 3, 4, 6].

Study Design and Patient Cohort

To clarify the clinical value of incidental findings on different imaging modalities, a retrospective cohort study was conducted within the Kaiser Permanente Southern California health system. The investigation focused on the diagnostic agreement between CT and MRI for covert cerebrovascular disease and the specific prognostic implications of findings from each. The study population consisted of 18,628 adults aged 50 years or older with no prior diagnosis of stroke or dementia who underwent both a head CT and a brain MRI within a 30-day window between 2009 and 2022. The cohort had a mean age of 64.9 years, and 59.1% were female. To analyze this large, real-world dataset, the researchers employed natural language processing, a computational method that extracts structured data from unstructured text, to identify mentions of covert brain infarction (CBI) and white matter disease (WMD) directly from the clinical radiology reports. The primary outcome was incident ischemic stroke or dementia over the follow-up period. The association between imaging findings and outcomes was assessed using Cox proportional hazards models, a statistical technique that estimates how various factors influence the risk of an event over time, after adjusting for age, sex, race, and vascular and dementia risk factors.

Modality Differences in Detecting Covert Cerebrovascular Disease

The analysis revealed important differences in how CT and MRI detect covert cerebrovascular disease. For covert brain infarction, the prevalence was nearly identical, identified in 6.3% of participants on CT and 6.1% on MRI. However, the agreement between the two modalities for identifying CBI in the same patient was modest, with a kappa statistic of 0.27, indicating frequent discordance. A much larger disparity appeared in the detection of white matter disease (WMD). MRI identified WMD in 60.5% of patients, more than double the 24.4% detection rate on CT. The agreement for WMD was similarly modest (κ = 0.23). Furthermore, when WMD was graded for severity, nearly half of the patients showed discordant results between the two scans. In these cases of disagreement, MRI assigned a higher severity grade than CT in 92% of instances. These results confirm that MRI is substantially more sensitive than CT for detecting WMD and characterizing its severity, a critical factor to consider when interpreting reports from both modalities.

Prognostic Implications: Risk Stratification by Imaging Modality

Beyond detection rates, the study's central finding lies in the differing prognostic value of WMD depending on the imaging modality that identified it. Over a mean follow-up of 4.4 years, the risk of incident ischemic stroke or dementia varied significantly. Patients with no WMD on either scan had a baseline incidence rate of 12.7 per 1,000 person-years (95% CI 11.5-14.0). This rate nearly doubled to 22.6 per 1,000 person-years (95% CI 21.0-24.2) for patients with WMD visible only on MRI. Strikingly, the rate was even higher for patients with WMD reported on CT only, at 37.0 per 1,000 person-years (95% CI 29.8-45.6). The highest risk was seen in patients with WMD detected on both modalities, with an incidence rate of 52.2 per 1,000 person-years (95% CI 48.7-56.0). After adjusting for confounders, the hazard ratios reinforced this pattern. Compared to patients with negative scans, those with MRI-only WMD had an adjusted hazard ratio of 1.23 (95% CI 1.07-1.41). For patients with WMD on both scans, the adjusted hazard ratio was 1.82 (1.58-2.11). The key clinical insight is that WMD detected on CT, even if missed or graded as less severe on a prior or subsequent MRI, identifies a patient subgroup with a substantially elevated risk of adverse outcomes.

Clinical Takeaways and Future Directions

This study provides a clear message for practicing clinicians: while MRI is more sensitive for detecting white matter disease, a finding of WMD on a CT scan carries significant negative prognostic weight. The data show that WMD visible on CT is associated with a higher risk for subsequent stroke or dementia than WMD that is only visible on MRI. Therefore, an incidental finding of WMD on a routine head CT should not be dismissed. It serves as a potent marker of underlying vascular pathology that warrants heightened clinical vigilance and aggressive management of modifiable vascular risk factors. This modality-specific interpretation is crucial for accurate patient risk stratification. The authors note several limitations, including the study's reliance on text from radiology reports rather than a direct, standardized review of the images themselves. Additionally, the cohort was composed of individuals who received both a CT and an MRI, a group that may not be representative of the general patient population. Future research may clarify if these findings apply broadly to patients who undergo only a single imaging study.

References

1. Scheuermann B, Parr SK, Schulze KM, et al. Associations of Cerebrovascular Regulation and Arterial Stiffness With Cerebral Small Vessel Disease: A Systematic Review and Meta‐Analysis. Journal of the American Heart Association. 2023. doi:10.1161/jaha.123.032616

2. Gorelick PB, Scuteri A, Black SE, et al. Vascular Contributions to Cognitive Impairment and Dementia. Stroke. 2011. doi:10.1161/str.0b013e3182299496

3. Morris Z, Whiteley W, Longstreth W, et al. Incidental findings on brain magnetic resonance imaging: systematic review and meta-analysis. BMJ. 2009. doi:10.1136/bmj.b3016

4. Kent DM, Leung LY, Zhou Y, et al. Association of Incidentally Discovered Covert Cerebrovascular Disease Identified Using Natural Language Processing and Future Dementia. Journal of the American Heart Association. 2022. doi:10.1161/jaha.122.027672

5. Blair GW, Appleton JP, Mhlanga I, et al. Design of trials in lacunar stroke and cerebral small vessel disease: review and experience with the LACunar Intervention Trial 2 (LACI-2). Stroke and Vascular Neurology. 2024. doi:10.1136/svn-2023-003022

6. Sharma B, Wang M, McCreary CR, Camicioli R, Smith EE. Gait and falls in cerebral small vessel disease: a systematic review and meta-analysis. Age and Ageing. 2023. doi:10.1093/ageing/afad011