Abstract
The study's objective was the evaluation of the diagnostic accuracy of the T2-FLAIR mismatch sign in terms of diagnosing IDH-mutant non-codeleted (IDHmut-Noncodel) lower grade gliomas (LGG) of the brain. We searched the MEDLINE, Scopus and Cochrane Central databases. The last database search was performed on 12 April 2021. Studies that met the following were included: MRI scan assessing the presence of T2-FLAIR mismatch sign, and available IDH mutation and 1p/19q codeletion status. The quality of studies was assessed using the QUADAS-2 tool. Twelve studies involving 14 cohorts were included in the quantitative analysis. The diagnostic odds ratio [DOR (95% confidence interval; CI)] was estimated at 34.42 (20.95, 56.56), P(z) < 0.01. Pooled sensitivity and specificity (95% CI) were estimated at 40% (31-50%; P(z) = 0.05) and 97% (93-99%; P(z) < 0.01), respectively. The likelihood ratio (LR; 95% CI) for a positive test was 11.39 (6.10, 21.29; P(z) < 0.01) and the LR (95% CI) for a negative test was 0.40 (0.24, 0.65; P(z) < 0.01).The T2-FLAIR mismatch sign is a highly specific biomarker for the diagnosis of IDHmut-Noncodel LGGs. However, the test was found positive in some other tumors and had a high number of false negative results. The diagnostic accuracy of the mismatch sign might be improved when combined with further imaging parameters.