Cognitive impairment in mild traumatic brain injury: a longitudinal diffusional kurtosis and perfusion imaging study

Abstract

Background and purpose: Cognitive impairment is frequent among patients with mild traumatic brain injury despite the absence of detectable damage on conventional MR imaging. In this study, the quantitative MR imaging techniques DTI, DKI, and ASL were used to measure changes in the structure and function in the thalamus and WM of patients with MTBI during a short follow-up period, to determine whether these techniques can be used to investigate relationships with cognitive performance and to predict outcome.

Materials and methods: Twenty patients with MTBI and 16 controls underwent MR imaging at 3T and a neuropsychological battery designed to yield measures for attention, concentration, executive functioning, memory, learning, and information processing. MK, FA, MD, and CBF were measured in the thalamus by using region-of-interest analysis and in WM by using tract-based spatial statistics. Analyses were performed comparing regional imaging measures of subject groups and the results of testing of their associations with neuropsychological performance.

Results: Patients with MTBI exhibited significant differences from controls for DTI, DKI, and ASL measures in the thalamus and various WM regions both within 1 month after injury and >9 months after injury. At baseline, DTI and DKI measures in the thalamus and various WM regions were significantly associated with performance in different neuropsychological domains, and cognitive impairment was significantly associated with MK in the thalamus and FA in optic radiations.

Conclusions: Combined application of DTI, DKI, and ASL to study MTBI might be useful for investigating dynamic changes in the thalamus and WM as well as cognitive impairment during a short follow-up period, though the small number of patients examined did not predict outcome.

Figures

Fig 1.

Plots displaying age- and sex-adjusted means (hash marks) and standard errors (lines) for MK, FA, MD, and CBF in the thalamus of controls, patients with MTBI at baseline visit, and patients with MTBI at follow-up visit. When patients with MTBI are compared with controls, they show significantly lower MK, FA, and CBF and higher MD at baseline visit (MK, FA, MD, and CBF: P < .01) and significantly lower MK and CBF at follow-up visit (MK and CBF: P < .01). All differences remain statistically significant after Bonferroni correction.

Fig 2.

Plots displaying age- and sex-adjusted means (hash marks) and standard errors (lines) for MK, FA, and MD in the total WM of controls, patients with MTBI at baseline visit, and patients with MTBI at follow-up visit. When patients with MTBI are compared with controls, they show significantly lower MK and FA and higher MD at baseline visit (MK, FA, and MD: P < .01) and significantly lower MK at follow-up visit (P < .01). All differences remain statistically significant after Bonferroni correction.

Fig 3.

Plots displaying age, sex, and time since injury adjusted means (hash marks) and standard errors (lines) for MK in the thalamus and FA in the optic radiations of controls and cognitively unimpaired patients with MTBI and cognitively impaired patients with MTBI at baseline visit. When cognitively impaired patients with MTBI are compared with cognitively unimpaired patients with MTBI, they show significantly lower MK in the thalamus (P < .01) and FA in the optic radiations (P = .04). Only the difference for MK in the thalamus remains significant after Bonferroni correction.