Abstract
PURPOSE: The purpose of this study was to examine the values of fractional anisotropy (FA) and apparent diffusion coefficient (ADC) in diffusion tensor imaging (DTI) for diagnosing patients with nerve impairment due to lumbar disc herniation (LDH). METHODS: A literature search of databases (PubMed, Web of Science, Cochrane Library and Embase) was systematically performed to identify articles published before September 2021 that were relevant to this study. FA and ADC estimates of compressed nerve roots due to LDH and healthy controls in the same segment were compared, with either fixed or random effects models selected according to I2 heterogeneity. Additionally, subgroup analysis, sensitivity analysis, potential publication bias analysis and meta-regression analysis were also performed. RESULTS: A total of 369 patients with LDH from 11 publications were included in this meta-analysis. The results showed significantly lower FA values (Weighted Mean Difference (WMD): -0.08, 95% confidence interval (CI): -0.09 to -0.07, P ≤ 0.001, I2 = 87.6%) and significantly higher ADC values (WMD: 0.25, 95% CI: 0.20 to 0.30, P ≤ 0.001, I2 = 71.4%) of the nerve on the compressed side due to LDH compared to the healthy side. Subgroup analysis indicated that different countries and magnetic field strengths may be associated with higher heterogeneity. Furthermore, meta-regression analysis further revealed that segment and field strength did not have a significant effect on the results, regardless of the FA or ADC values. Contrastingly, in FA, the year of publication, country, b value and directions showed an effect on the results. CONCLUSIONS: This meta-analysis showed a significant decrease in FA and a significant increase in ADC in patients with nerve damage due to LDH. The results favourably support the presence of nerve impairment in patients with LDH.