Gait apraxia evaluation in normal pressure hydrocephalus using inertial sensors. Clinical correlates, ventriculoperitoneal shunt outcomes, and tap-test predictive capacity

Background Idiopathic normal pressure hydrocephalus (iNPH) is a neurological condition with gait apraxia signs from its early manifestation. Ventriculoperitoneal shunt (VPS) is a surgical procedure available for treatment. The Cerebrospinal fluid Tap Test (CSF-TT) is a quick test used as selection criterion for VPS treatment. Its predictive capacity for VPS outcomes is still sub judice. This study is aimed to test the hypothesis that wearable motion sensors provide valid measures to manage iNPH patients with gait apraxia. Methods Forty-two participants of the Bologna PRO-Hydro observational cohort study were included in the analyses. The participants performed the Timed Up and Go (TUG) and the 18 m walking test (18mW) with inertial sensors at baseline, three days after the CSF-TT, and six months after VPS. 21 instrumental variables described gait and postural transitions from TUG and 18mW recordings. Furthermore, participants were clinically assessed with scales (clinical variables). We tested the hypothesis by analysing the concurrent validity of instrumental and clinical variables, their individual- and group-level responsiveness to VPS, and their predictive validity for VPS outcomes after CSF-TT. Results The instrumental variables showed moderate to high correlation with the clinical variables. After VPS, most clinical and instrumental variables showed statistically significant improvements that reflect a reduction of apraxic features of gait. Most instrumental variables, but only one clinical variable (i.e., Tinetti POMA), had predictive value for VPS outcomes (significant adjusted R-2 in the range 0.12-0.70). Conclusions These results confirm that wearable inertial sensors may represent a valid tool to complement clinical evaluation for iNPH assessment and prognosis.