Mild cognitive impairment in spontaneous intracranial hypotension and its rapid reversal by repair of a spinal cerebrospinal fluid leak.

Background: Patients with spontaneous intracranial hypotension (SIH) report difficulties in concentration and memory. To objectify these deficits, we implemented standard cognitive tests into our routine SIH workup.

Method: Retrospective, single-center report of cognitive standard tests among patients with SIH consecutively admitted from May to July 2023. Cognitive testing involved the Montreal Cognitive Assessment (MoCA©, alternate versions, 0-30 points, 30 points for best performance, ≤26 indicating mild cognitive impairment at age >64 years), and the Trail Making Test, part B (TMT B, z-scores adjusted to age and education) to test for executive function. Both were administered at admission, and within 36-72 h after surgical repair of the spinal cerebrospinal fluid (CSF) leak.

Results: A total of 18 patients with an active spinal CSF leak were tested at admission (seven with ventral, three with lateral leak, and eight with CSF-venous fistula). There was no profound brain sagging as described in brain sagging dementia. The mean (standard deviation [SD]) age was 53.6 (11) years. Bern scores ranged between 0 and 9, median 6.5. The mean (SD) MoCA score at admission was 26.5 (2) points, with five patients (28%) scoring <26 points indicative of mild cognitive impairment. Performance in the TMT B was impaired in nine patients (50%, z-score ≥2). Upon targeted treatment of the CSF leak, the mean (SD) MoCA score immediately improved to 28.5 (1), p = 0.001 (n = 14), as did performance on the TMT B (mean [SD] 2.1 [2] vs. 1.1 [1], p = 0.015, n = 13).

Discussion: Spontaneous intracranial hypotension with an active spinal CSF leak is associated with cognitive impairment and surgical closure of the leak led to rapid improvement. We conclude that there may be a causal relationship between cognitive dysfunction and spinal CSF loss. We suggest considering spinal CSF leaks as a treatable cause in patients with mild cognitive impairment and with pre-dementia. This may ultimately necessitate thorough screening of brain and spine magnetic resonance images in patients with mild cognitive impairment.