Systematic Review and Meta-Analyses of Word Production Abilities in Dysfunction of the Basal Ganglia: Stroke, Small Vessel Disease, Parkinson's Disease, and Huntington's Disease

Clinical populations with basal ganglia pathologies may present with language production impairments, which are often described in combination with comprehension measures or attributed to motor, memory, or processing-speed problems. In this systematic review and meta-analysis, we studied word production in four (vascular and non-vascular) pathologies of the basal ganglia: stroke affecting the basal ganglia, small vessel disease, Parkinson's disease, and Huntington's disease. We compared scores of these clinical populations with those of matched cognitively unimpaired adults on four well-established production tasks, namely picture naming, category fluency, letter fluency, and past-tense verb inflection. We conducted a systematic search in PubMed and PsycINFO with terms for basal ganglia structures, basal ganglia disorders and language production tasks. A total of 114 studies were included, containing results for one or more of the tasks of interest. For each pathology and task combination, effect sizes (Hedges' g) were extracted comparing patient versus control groups. For all four populations, performance was consistently worse than that of cognitively unimpaired adults across the four language production tasks (p-values < 0.010). Given that performance in picture naming and verb inflection across all pathologies was quantified in terms of accuracy, our results suggest that production impairments cannot be fully explained by motor or processing-speed deficits. Our review shows that while language production difficulties in these clinical populations are not negligible, more evidence is necessary to determine the exact mechanism that leads to these deficits and whether this mechanism is the same across different pathologies.

Associations of Particulate Matter Exposures With Brain Gray Matter Thickness and White Matter Hyperintensities: Effect Modification by Low-Grade Chronic Inflammation

BACKGROUND

Numerous studies have shown the effect of particulate matter exposure on brain imaging markers. However, little evidence exists about whether the effect differs by the level of low-grade chronic systemic inflammation. We investigated whether the level of c-reactive protein (CRP, a marker of systemic inflammation) modifies the associations of particulate matter exposures with brain cortical gray matter thickness and white matter hyperintensities (WMH).

METHODS

We conducted a cross-sectional study of baseline data from a prospective cohort study including adults with no dementia or stroke. Long-term concentrations of particulate matter ≤ 10 µm in diameter (PM10) and ≤ 2.5 µm (PM2.5) at each participant's home address were estimated. Global cortical thickness (n = 874) and WMH volumes (n = 397) were estimated from brain magnetic resonance images. We built linear and logistic regression models for cortical thickness and WMH volumes (higher versus lower than median), respectively. Significance of difference in the association between the CRP group (higher versus lower than median) was expressed as P for interaction.

RESULTS

Particulate matter exposures were significantly associated with a reduced global cortical thickness only in the higher CRP group among men (P for interaction = 0.015 for PM10 and 0.006 for PM2.5). A 10 μg/m³ increase in PM10 was associated with the higher volumes of total WMH (odds ratio, 1.78; 95% confidence interval, 1.07-2.97) and periventricular WMH (2.00; 1.20-3.33). A 1 μg/m³ increase in PM2.5 was associated with the higher volume of periventricular WMH (odds ratio, 1.66; 95% confidence interval, 1.08-2.56). These associations did not significantly differ by the level of high sensitivity CRP.

CONCLUSION

Particulate matter exposures were associated with a reduced global cortical thickness in men with a high level of chronic inflammation. Men with a high level of chronic inflammation may be susceptible to cortical atrophy attributable to particulate matter exposures.

A Positive Association between the Atherogenic Index of Plasma and White Matter Hyperintensity

BACKGROUND

White matter hyperintensity (WMH) is a risk factor for dementia and ischemic stroke. The atherogenic index of plasma (AIP) is a simple and cost-effective marker for the prediction of various vascular diseases. In this study, we evaluated the relationship between AIP and WMH in adults without cerebrovascular accidents.

METHODS

We analyzed the data of 281 adults, aged ≥26 years, who underwent brain magnetic resonance imaging (MRI) at the health promotion center of an education hospital between January 2014 and December 2018. Participants were divided into three categories according to tertiles of the AIP scores (T1: <0.20; T2: 0.20-0.48; and T3: >0.48). WMH was defined as a modified Fazekas scale score of 1-3 on brain MRI. A cubic spline curve was used to determine the linearity of the relationship between AIP and WMH. Multiple logistic regression analysis was used to evaluate the relationship between the AIP and WMH.

RESULTS

The prevalence of WMH was 45.7% in T1, 57.0% in T2, and 66.0% in T3 (T3 vs. T1, P for post-hoc analysis=0.005). The increased odds of WMH were associated with increased AIP. The odds ratio (OR) with a 95% confidence interval (CI) for WMH of T2 and T3 compared with T1 were 1.57 (0.88-2.80) and 2.30 (1.28-4.14), respectively. After adjusting for confounding variables, the OR with a 95% CI for WMH in the T2 and T3 groups vs. the referent T1 were 1.55 (0.76-3.13) and 2.27 (1.06-4.84), respectively.

CONCLUSION

AIP is independently and positively associated with WMH in a healthy population.

Neuropsychiatric systemic lupus erythematosus is associated with a distinct type and shape of cerebral white matter hyperintensities

Objectives

Advanced white matter hyperintensity (WMH) markers on brain MRI may help reveal underlying mechanisms and aid in the diagnosis of different phenotypes of SLE patients experiencing neuropsychiatric (NP) manifestations.

Methods

In this prospective cohort study, we included a clinically well-defined cohort of 155 patients consisting of 38 patients with NPSLE (26 inflammatory and 12 ischaemic phenotype) and 117 non-NPSLE patients. Differences in 3 T MRI WMH markers (volume, type and shape) were compared between patients with NPSLE and non-NPSLE and between patients with inflammatory and ischaemic NPSLE by linear and logistic regression analyses corrected for age, sex and intracranial volume.

Results

Compared with non-NPSLE [92% female; mean age 42 (13) years], patients with NPSLE [87% female; mean age 40 (14) years] showed a higher total WMH volume [B (95%-CI)]: 0.46 (0.0 7 ↔ 0.86); P = 0.021], a higher periventricular/confluent WMH volume [0.46 (0.0 6 ↔ 0.86); P = 0.024], a higher occurrence of periventricular with deep WMH type [0.32 (0.1 3 ↔ 0.77); P = 0.011], a higher number of deep WMH lesions [3.06 (1.2 1 ↔ 4.90); P = 0.001] and a more complex WMH shape [convexity: ‒0.07 (‒0.12 ↔ ‒0.02); P = 0.011, concavity index: 0.05 (0.0 1 ↔ 0.08); P = 0.007]. WMH shape was more complex in inflammatory NPSLE patients [89% female; mean age 39 (15) years] compared with patients with the ischaemic phenotype [83% female; mean age 41 (11) years] [concavity index: 0.08 (0.0 1 ↔ 0.15); P = 0.034].

Conclusion

We demonstrated that patients with NPSLE showed a higher periventricular/confluent WMH volume and more complex shape of WMH compared with non-NPSLE patients. This finding was particularly significant in inflammatory NPLSE patients, suggesting different or more severe underlying pathophysiological abnormalities.

ROBUST WHITE MATTER HYPERINTENSITY SEGMENTATION ON UNSEEN DOMAIN

Typical machine learning frameworks heavily rely on an underlying assumption that training and test data follow the same distribution. In medical imaging which increasingly begun acquiring datasets from multiple sites or scanners, this identical distribution assumption often fails to hold due to systematic variability induced by site or scanner dependent factors. Therefore, we cannot simply expect a model trained on a given dataset to consistently work well, or generalize, on a dataset from another distribution. In this work, we address this problem, investigating the application of machine learning models to unseen medical imaging data. Specifically, we consider the challenging case of Domain Generalization (DG) where we train a model without any knowledge about the testing distribution. That is, we train on samples from a set of distributions (sources) and test on samples from a new, unseen distribution (target). We focus on the task of white matter hyperintensity (WMH) prediction using the multi-site WMH Segmentation Challenge dataset and our local in-house dataset. We identify how two mechanically distinct DG approaches, namely domain adversarial learning and mix-up, have theoretical synergy. Then, we show drastic improvements of WMH prediction on an unseen target domain.