Relative cerebral blood volume and fractional anisotropy for differentiating recurrent glioma from radiation injury: a systematic review

Background: Glioma recurrence usually indicates a poorer prognosis and impacts survival prediction, while radiation necrosis may be reversible or non-progressive. To address this challenge and guide appropriate clinical decisions, advanced imaging modalities and biomarkers are being used to non-invasively assess lesion characteristics and improve diagnostic accuracy. This advanced imaging modalities include relative cerebral blood volume (rCBV) in perfusion magnetic resonance imaging (MRI) and fractional anisotropy (FA) in diffusion tensor imaging (DTI). However, the value of rCBV and FA in evaluation of glioma recurrence and radiation injury was rarely studied. The aim is to explore the value of rCBV and FA towards differentiating recurrent glioma from radiation injury.

Methods: Published studies related to the perfusion MRI and DTI of recurrent glioma and radiation injury were retrieved from PubMed, Cochrane Library and Web of Science database before 1 September 2025. Data extraction results were managed using Excel for a meta-analysis. Newcastle-Ottawa scale was used to assess the quality of included studies. Pooling data synthesis was completed using the Review Manager 5.4 software with a random effect model.

Results: Eighteen qualified studies were used for qualitative and quantitative data synthesis. The rCBV was higher in the recurrent glioma group than the one in the radiation injury group [standardized mean difference (SMD) =1.48, 95% confidence interval (CI): 1.13, 1.84, P<0.00001, I²=74%]. The FA was higher in recurrent glioma group than the one in radiation injury group (SMD =1.13, 95% CI: 0.52, 1.73, and P=0.0003, I²=72%). In our study, the average threshold of rCBV in recurrent glioma group and radiation injury group was respectively 3.896±1.704 and 1.416±0.648 based on the rCBV results of 16 studies.

Conclusions: The well-defined search strategy, together with the rigorous study selection process and methodological quality assessment, strengthens the validity of this meta-analysis. Our findings suggest that rCBV and FA have potential value in distinguishing recurrent glioma from radiation injury. However, given the heterogeneity among studies, further well-designed prospective studies are required to validate these findings.