Distinguishing between low-grade and high-grade brainstem glioma using standard MRI pulse sequences

Abstract

Purpose: This retrospective study employs a quantitative analysis of signal intensities derived from standard MRI pulse sequences to differentiate between low-grade and high-grade brainstem gliomas (BSGs).

Materials and methods: Forty-three patients with histopathologically confirmed BSGs underwent gadolinium-enhanced brain MRI. Quantitative parameters, including mean, median, standard deviation, maximum, minimum, and lesion-to-normal tissue ratios, were extracted from volumes of interest (VOIs) placed on pre- and post-contrast T1-weighted, fluid-attenuated inversion recovery (FLAIR), and apparent diffusion coefficient (ADC) maps. Receiver operating characteristic curve analysis was performed to assess the diagnostic performance of each parameter.

Result: Quantitative analysis of T1-weighted and FLAIR sequences revealed that mean T1 signal intensity (T1_mean), median T1 signal intensity (T1_median), and minimum FLAIR signal intensity (FLAIR_min) were significant discriminators between low-grade and high-grade BSGs. Optimal cut-off values, sensitivities, and specificities for these parameters were as follows: 559.5 (87.5%, 65.5%) for T1_mean, 576.5 (78.6%, 69%) for T1_median, and 349 (79.3%, 64.3%) for FLAIR_min. Analysis of the solid tumor component on ADC maps identified minimum ADC (ADCs_min) and the ratio of mean ADC to normal white matter ADC (rADCs_mean) as significant discriminators, with optimal cut-off values, sensitivities, and specificities of 862.5 x 10⁻⁶ mm²/s (42.9%, 93.1%) and 1.4785 (92.9%, 48.3%), respectively.

Conclusion: Quantitative signal intensity analysis of conventional MRI sequences, particularly T1-weighted and FLAIR, can effectively differentiate between low-grade and high-grade brainstem gliomas (BSGs). Furthermore, analysis of the solid tumor component on ADC maps provides valuable discriminatory information.