# Value of biomarkers in the prediction of shunt responsivity in patients with normal pressure hydrocephalus

## TL;DR

This study investigates whether biomarkers in cerebrospinal fluid can predict which patients with normal pressure hydrocephalus will benefit from shunt surgery, but finds clinical factors are more reliable.

## Contribution

The study evaluates the predictive value of CSF biomarkers for shunt responsiveness in NPH and finds they do not improve prediction over clinical factors.

## Key findings

- CSF biomarkers like NfL, NfH, and Aβ42 showed no significant differences between shunt responders and non-responders.
- Clinical factors (e.g., iNPH scale) had better predictive accuracy than individual biomarkers.
- Combining biomarkers with clinical factors did not enhance predictive accuracy.

## Abstract

Preoperative differentiation between responders and non-responders to ventriculoperitoneal (VP) shunting in the treatment of normal pressure hydrocephalus (NPH) remains a significant challenge. Identifying biomarkers in presurgical assessment represents a promising approach to reducing the need for invasive cerebrospinal fluid CSF testing. In this prospective observational study, thirty adult patients were classified into Group A (responders to VP shunting) and Group B (non-responders) based on their responsiveness to invasive CSF testing. The overall clinical condition and Idiopathic NPH (iNPH) scale were assessed at baseline. Additionally, biomarker levels were compared between the two groups. Elevated levels of Neurofilament Light Chain (NfL) and Neurofilament Heavy Chain (NfH) in CSF and a reduced level of beta-amyloid Aβ42 were observed. No significant differences in biomarker levels were found between groups. Individual biomarkers demonstrated only poor predictive value (AUC = 0.37–0.53). Clinical factors were stronger predictors (AUC = 0.642–0.669), with no improvement when combined with all examined biomarkers (AUC = 0.428–0.431). No single biomarker reliably predicted confirmed postoperative shunt responsiveness among patients who underwent VP shunt placement and demonstrated clinical improvement. Clinical factors were stronger predictors, suggesting that patient history and clinical assessment (e.g., the iNPH scale) provide more reliable diagnostic information. Notably, combining biomarkers with clinical factors did not improve predictive accuracy.