Diffuse pachymeningeal T2-FLAIR hyperintensity in spontaneous intracranial hypotension: to obviate gadolinium administration or not

Excerpt

I read with great interest the article by O’Cearbhaill et al. [1] on the potential clinical impact that finding a pachymeningeal thickening or hyperintensity on non-contrast fluid-attenuated inversion recovery (FLAIR) sequence might have in patients with spontaneous intracranial hypotension (SIH), specifically after linking it to the well-known finding of diffuse pachymeningeal thickening and enhancement on post-contrast T1-weighted imaging (T1WI). All patients with SIH in their sample (n = 26) presented both of the aforementioned findings, something that, as suggested by the authors, might have two main advantages: (1) to obviate the need of gadolinium administration and (2) to perform a simpler and faster MRI protocol for the evaluation of typical intracranial SIH findings. This concept had already been explored a few years ago by Tosaka et al. [2], based in a lower sample of patients (n = 8); they had, however, the advantage of presenting MRI studies performed in 3 different chronological occasions: initial (4 days to 2 months after onset of headache), second (10 days to 2 months after initial study), and follow-up (3 to 5 months after initial study). Although their main conclusion was, much like the one by O’Cearbhaill et al. [1], that diffuse pachymeningeal hyperintensity on FLAIR might be a similar sign to diffuse pachymeningeal enhancement on T1WI, it was also reported a discrepancy in the resolution of such findings, with the former disappearing more rapidly than the latter. The hypotheses raised by the authors were (1) that it could be due to different sensitivity of both sequences to such findings (i.e., FLAIR less sensitive) or (2) the relative rates of response to normalization of intracranial pressure after treatment. Given the low sample of both studies, sensitivity and specificity analysis were not possible and, as such, I wonder if the potential least sensitivity of the FLAIR sequence outweighs the advantage of avoiding gadolinium administration. Moreover, as the authors state, the potential lack of sensitivity might be boosted by the widespread use of FLAIR 3D over 2D sequences. Due to this potential pitfall and to the well-validated Bern score [3] for the prediction of spinal cerebrospinal fluid (CSF) leaks in patients with SIH (which uses pachymeningeal enhancement as a major imaging finding—2 points), it seems premature to obviate contrast agent administration per protocol at this point. …