Idiopathic Normal Pressure Hydrocephalus: multiparametric study of a cohort of patients undergoing cerebrospinal fluid tap test.

Idiopathic Normal Pressure Hydrocephalus (iNPH) is a significant cause of reversible dementia in the elderly, marked by gait disturbances, cognitive impairment, and urinary symptoms, accompanied by ventriculomegaly on brain imaging. The diagnostic process for iNPH is challenging due to symptom overlap with other degenerative neurological diseases. The current diagnostic guidelines involve symptomatic and radiological findings, while treatment typi- cally involves ventriculoperitoneal or lumboperitoneal shunts [1]. Given the risks associated with these invasive procedures, especially in older patients, careful selection is crucial [2]. The iNPH Radscale is a validated, standardized neuroimaging assessment tool used to evaluate and quantify radiological features associated with iNPH, which help in diagnosing iNPH and predicting response to shunt- ing. The removal of cerebrospinal fluid through a lumbar puncture (Tap Test, TT) is commonly employed to select candidates for shunt surgery [3]. It has a high positive predictive value but low sensitivity and negative predictive value. Accurately diagnosing iNPH is crucial, as up to 90% of patients improve after ventriculoperitoneal shunting. Therefore, refining diagnostic procedures is essential to distinguish suitable surgical candidates from those with other mimicking conditions or comorbidities that affect treatment outcomes. We aimed to characterize a cohort of patients with suspected iNPH undergoing the TT, to identify parameters that might help differentiate TT-negative patients who could be false negatives and might benefit from personalized management. Materials and Methods: Patients aged 60-85 years with suspected iNPH, admitted for TT evaluation. The cohort was retrospectively and prospectively studied, focusing on symptoms and neuroimaging. Comprehensive cognitive and motor evaluations were performed using standardized tests (MMSE, MOCA, FAB, SDMT, TMT-A), and neuroimaging features were assessed using the iNPH Radscale, Global Cortical Atrophy Scale, and Fazekas Scale for WMLs. Results: We studied 32 patients, average age 74 years, including 14 females. 21 patients did not respond to the TT, while 10 showed improvement. Responders had significantly higher Radscale scores (p=0.0001), and lower cortical atrophy (p=0.03) compared to non- responders. Specific Radscale components (Evans index ≥ 0.3, tight sulci, focal sulcal dilation, temporal horns, callosal angle) were more prevalent among responders. ROC analysis suggested cut-off levels of 6 for Radscale score and 0.35 for Evans index to differentiate responders from non-responders. Conclusions: The findings suggest that a multidimensional evaluation, incorporating standardized neuroradiological assessments with the Radscale and cortical atrophy measures, is crucial in identifying potential false-negative TT responses. This approach could guide more stringent follow-up and potential repeat TT evaluations for non-responders. The findings need confirmation in larger, prospective studies.