18F-FDG brain PET: a metabolic predictive factor for gait improvement after cerebrospinal fluid shunting in normal pressure hydrocephalus?

The pathophysiology of normal pressure hydrocephalus (NPH) has not been fully elucidated. Treating NPH with cerebrospinal fluid shunts to improve gait disturbances may have some risks and inconsistent benefits. No clear predictive factor has been identified thus far. This preliminary study aimed to evaluate the predictive value of preoperative brain 18F-FDG positron emission tomography (PET) on overall gait response in patients with NPH. Sixteen patients with NPH who underwent 18F-FDG PET before shunt surgery between 2012 and 2022 were included retrospectively and separated into two groups based on their gait response one year after surgery: responders (R) or nonresponders (NR). Brain glucose metabolism was assessed using visual and semiquantitative analyses using SPM8 software (Welcome Department of Cognitive Neurology, University College, London, UK). Five regions of interest were selected: global cortex, cerebellum, thalamus, striatum, and midbrain. Visual interpretation showed more frequent hypometabolism of the striatum, thalamus and global cortex in NR. None of the patients showing hypometabolism of these regions were R. Based on these results, the visual interpretation allowed us to identify 3/8 NR and 8/8 R. Semiquantitative analysis confirmed significantly lower thalamic metabolism in the NR group (P=0.037) and a trend towards lower metabolism of the striatum (P=0.075) with an area under the curve of 0.77 for thalamic metabolism to discriminate between R and NR. This preliminary study using brain 18F-FDG PET suggests that reduced brain metabolism in the thalamus and striatum along with cortical hypometabolism may be associated with poorer gait response to CSF shunting in normal pressure hydrocephalus (NPH). Although these findings suggest that preoperative brain 18F-FDG PET could potentially aid in selecting appropriate candidates for shunt surgery, further research with larger sample sizes is needed to confirm these results.