Application of diffusion tensor imaging in the diagnosis of post-stroke aphasia: a meta-analysis and systematic review.

Introduction: Diffusion Tensor Imaging (DTI) indicators of different white matter (WM) fibers and brain region lesions for post-stroke aphasia (PSA) are inconsistent in existing studies. Our study examines the consistency and differences between PSA tests performed with DTI. In addition, obtaining consistent and independent conclusions between studies was made possible by utilizing DTI in PSA assessment.

Methods: In order to gather relevant studies using DTI for diagnosing PSA, we searched the Web of Science, PubMed, Embase, and CNKI databases. Based on the screening and evaluation of the included studies, the meta-analysis was used to conduct a quantitative analysis. Narrative descriptions were provided for studies that met the inclusion criteria but lacked data.

Results: First, we reported on the left hemisphere. The meta-analysis showed that fractional anisotropy (FA) of the arcuate fasciculus (AF) and superior longitudinal fasciculus (SLF), inferior frontal-occipital fasciculus (IFOF), inferior longitudinal fasciculus (ILF), and uncinate fasciculus (UF) were decreased in the PSA group in comparison with the healthy controls ( < 0.00001). However, in the comparison of axial diffusivity (AD), there was no statistically significant difference in white matter fiber tracts in the dual-stream language model of the PSA group. Elevated radial diffusivity (RD) was seen only in the IFOF and ILF ( = 0.01;  = 0.05). In the classic Broca's area, the FA of the PSA group was decreased ( < 0.00001) while the apparent diffusion coefficient was elevated ( = 0.03). Secondly, we evaluated the white matter fiber tracts in the dual-stream language model of the right hemisphere. The FA of the PSA group was decreased only in the IFOF ( = 0.001). AD was elevated in the AF and UF ( < 0.00001; PUF = 0.009). RD was elevated in the AF and UF ( = 0.01;  = 0.003). The other fiber tracts did not undergo similar alterations.

Conclusion: In conclusion, DTI is vital for diagnosing PSA because it detects WM changes effectively, but it still has some limitations. Due to a lack of relevant language scales and clinical manifestations, diagnosing and differentiating PSA independently remain challenging.

Systematic Review Registration: https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=365897.