Idiopathic normal pressure hydrocephalus and frontotemporal dementia: an unexpected association

Abstract

Idiopathic normal pressure hydrocephalus has a complex multifactorial pathogenesis and is associated with Alzheimer's disease in many patients. To date, it is not well known if a similar association exists with behavioural variant of frontotemporal lobar degeneration. In a first step, we compare the prevalence of idiopathic normal pressure hydrocephalus in two groups of patients, one with behavioural variant of frontotemporal lobar degeneration (n = 69) and the other with Alzheimer's disease (n = 178). In the second step, we describe more precisely the phenotype of patients with the association of idiopathic normal pressure hydrocephalus and behavioural variant of frontotemporal lobar degeneration. Firstly, we report that the prevalence of idiopathic normal pressure hydrocephalus was far higher in the group of patients with behavioural variant of frontotemporal lobar degeneration than in the group of patients with Alzheimer's disease (7.25% and 1.1%, respectively, P = 0.02). Secondly, we show that patients with the double diagnosis share common clinical and para-clinical features of both idiopathic normal pressure hydrocephalus and behavioural variant of frontotemporal lobar degeneration patients, including CSF shunting efficacy in real-life experience. Overall, our results suggest a link between these two conditions and should encourage neurologists to look for idiopathic normal pressure hydrocephalus in their behavioural variant of frontotemporal lobar degeneration patients in the event of gait disturbances; the benefit/risk balance could indeed be in favour of shunt surgery for selected patients with this newly described entity.

Keywords: Alzheimer’s disease; behavioural variant of frontotemporal lobar degeneration; idiopathic normal pressure hydrocephalus.

Graphical abstract

Figure 1

Flow chart. iNPH was diagnosed in 5 (7.25%) of the 69 bv-FTLD patients and in 2 (1.1%) of the 178 Alzheimer’s disease patients seen in 2019 (P = 0.02, Fisher’s exact test)

Figure 2

Imaging features of iNPH-bv-FTLD patients. (A) Axial T2-FLAIR MR image of an iNPH-bv-FTLD patient, showing frontotemporal asymmetrical atrophy, with a right predominance and ventriculomegaly with two segments (red arrows) used to calculate the Evans index. (B) Evans index for all patients of the three groups. (C) Coronal T2-FLAIR MR image of the same patient, showing a sharp callosal angle (red lines) and a convexity of the third ventricular walls. (D) Callosal angle (degrees) for all patients of the three groups. (E–G) Other T2-FLAIR MR images of the same iNPH-bv-FTLD patient: focal dilated sulci ((E), arrows), tight superior sulci at the convexity (F), cingulate sulcus sign ((G), arrow). (H) Example of axial slices of 18F-FDG PET scans in one patient, displayed with a rainbow colour scale, after metabolic normalization to the putamen. They show a diffuse hypometabolism of the subcortical and striatal structures that might be secondary to the iNPH, although the anterior and right predominance is consistent with an association with bv-FTLD. (I) As an example of basal ganglia hypometabolism, quantification of mean left caudate activity is shown for patients who had FDG-PET imaging. The values are normalized to the mean value of the bv-FTLD group. **P < 0.01 (Mann–Whitney test, U = 1). On each graph, the red symbol corresponds to the quantitative value of the illustrative patient

Figure 3

CSF Aβ₄₂ (A), T-Tau (B) and P-Tau (C) in iNPH-bv-FTLD (n = 7/9), bv-FTLD (n = 6/9) and iNPH (n = 4/9) groups. To compare measures made at different periods, the CSF biomarker levels were expressed relative to the cut-offs applied at the moment of the collection. One T-Tau value from the iNPH-bv-FTLD group was excluded as it was identified as an outlier. iNPH-bv-DLFT versus bv-DLFT (Mann–Whitney test): (A) ** P < 0.01 (U = 0); (B) *P < 0.05 (U = 4); (C) *P < 0.05 (U = 4.5)

Figure 4

Flow-chart presenting the effect of tap test (subtractive lumbar punctures) and shunt on gait in patients in the iNPH-bv-FTLD and iNPH groups. The effect on gait was assessed 3 months after shunt surgery. LP, lumbar puncture; ALS, amyotrophic lateral sclerosis; VPS, ventriculoperitoneal shunt; VAS, ventriculoatrial shunt