Reduced cerebellar cortical thickness in World Trade Center responders with cognitive impairment

Abstract

Prior research has demonstrated high levels of cognitive and physical functional impairments in World Trade Center (WTC) responders. A follow-up neuroimaging study identified changes to white matter connectivity within the cerebellum in responders with cognitive impairment (CI). In the first study to examine cerebellar cortical thickness in WTC responders with CI, we fielded a structural magnetic resonance imaging protocol. WTC responders (N = 99) participated in a structural magnetic resonance imaging (MRI) study, of whom 48 had CI. Participants with CI did not differ demographically or by intracranial volume when compared to cognitively unimpaired participants. MRIs were processed using the CERES imaging pipeline; bilateral cortical thickness in 12 cerebellar lobules was reported. Analyses were completed comparing mean cerebellar cortical thickness across groups. Lobules were examined to determine the location and functional correlates of reduced cerebellar cortical thickness. Multivariable-adjusted analyses accounted for the false discovery rate. Mean cerebellar cortical thickness was reduced by 0.17 mm in responders with CI. Decrements in cerebellar cortical thickness were symmetric and located in the Cerebellar Crus (I and II), and in Lobules IV, VI, VIIb, VIIIa, VIIIb, and IX. Cerebellar cortical thickness was associated with episodic memory, response speed, and tandem balance. WTC responders with CI had evidence of reduced cerebellar cortical thickness that was present across lobules in a pattern unique to this cohort.

Fig. 1. Cerebellar cortical thickness map.

Example of sagittal and coronal cerebellar cortical thickness estimate maps employed in this present study using a rainbow colormap to visually identify regional asymmetry in cortical thickness, such as in the cerebellar crus in this example of a responder with Cognitive Impairment.

Fig. 2. Reductions in mean bilateral cortical thickness (mm) in the whole cerebellum and across 12 cerebellar regions of interest in World Trade Center responders with cognitive impairment (n = 48) when compared to WTC responders who were cognitively unimpaired (n = 51).

Confidence intervals are shown using error bars. Reductions with confidence intervals not crossing the base line were nominally significant, while those marked with * survived adjustment for the false discovery rate.

Fig. 3. Association between cerebellar cortical thickness (mm) and episodic memory (coefficient = 0.35, standard error = 0.15, p-value = 0.021) adjusted for gender.

Males: small red circles and solid green line; females: large blue squares and dashed line; 95% confidence intervals shown using transparent gray boxes.