Larger hippocampal and brainstem volumes in high volume lumbar puncture responders as a predictor of shunt response in normal pressure hydrocephalus

Abstract

AbstractBackgroundNormal pressure hydrocephalus (NPH) is a treatable cause of dementia, urinary incontinence and gait abnormality which is important to diagnose early as the longer the delay in treatment, the worse the outcome after shunting. Gait improvement following high volume lumbar puncture (HVLP) and continuous lumbar drain (cLD) is widely used to predict shunt responsiveness in patients with suspected NPH. Here, we investigate differences in MRI volumetric and traditional measures between HVLP/cLD responders and non‐responders to identify imaging features that may help predict shunt response.Method82 patients with suspected NPH were studied retrospectively. Gait testing was performed before and immediately/24‐hours/72‐hours after CSF removal. A positive response to HVLP/cLD was defined as improvement in gait following the procedure. The 36 responders (26 men; mean‐age 79.3±6.3) and 46 non‐responders (25 men; mean‐age 77.2±6.1) underwent pre‐procedure 3T brain MRI including 3D‐MPRAGE sequence. 66 subcortical regional volumes were segmented using FreeSurfer. After normalizing for total intracranial volume, two‐tailed student's t‐test and chi‐square test were used to characterize statistical group‐differences. Assessment of Evan’s‐index (EI), callosal‐angle (CA) and disproportionately enlarged subarachnoid space hydrocephalus (DESH) were also made. Multivariable logistic regression models were tested using Akaike information criterion to determine which combination of metrics is most accurate in prediction of HVLP/cLD response.ResultResponders and non‐responders demonstrated no differences in total ventricular and white/gray matter volumes and in EI and DESH. CA (in men) and third and fourth ventricular volumes were decreased; and hippocampal and brainstem volumes were increased in responders (p<0.05). Temporal horns volume correlated with degree of improvement in gait velocity in responders (p=0.0006). The regression model was 76.8% accurate for HVLP/cLD response with AUC of 0.81.ConclusionImaging biomarkers, such as CA, third and fourth ventricular volumes and hippocampal and brainstem volumes, combined in a predictive model, may provide a useful noninvasive diagnostic supportive tool potentially helpful in (a) replacing the need for HVLP/cLD, and (b) predicting the success of shunt surgery. Larger hippocampi and brainstem in responders may reflect decreased rate of other comorbidities that contribute to hippocampal and brainstem atrophy, such as possible concomitant pathology of Alzheimer's disease, in responders as compared to non‐responders.