# Change in Optic Nerve Sheath Diameter and Cerebral Ventricular Shunt Failure in Children

## TL;DR

A change in optic nerve sheath diameter measured by ultrasound may help identify children with shunt failure in emergency settings.

## Contribution

This study introduces a noninvasive method using ocular ultrasound to screen for shunt failure in children with hydrocephalus.

## Key findings

- A change of 0.4 mm or more in optic nerve sheath diameter was associated with shunt failure.
- A change below 0.4 mm had a 98% negative predictive value for shunt failure.
- The area under the ROC curve was 0.86, indicating strong diagnostic accuracy.

## Abstract

Is a change in optic nerve sheath diameter measured on ocular point-of-care ultrasonography associated with shunt failure among symptomatic children with cerebrospinal fluid shunts?

This cohort study found that a change of 0.4 mm or more in optic nerve sheath diameter from prior asymptomatic baseline was associated with shunt failure in symptomatic children. A change below this cutoff had a 98% negative predictive value for shunt failure.

This study suggests that a change in the optic nerve sheath diameter from a patient’s own asymptomatic baseline measured by ocular point-of-care ultrasonography may be a helpful noninvasive screening tool for shunt failure among symptomatic children presenting to the emergency department.

Ocular point-of-care ultrasonography (POCUS) may be a clinically useful method to evaluate shunt dysfunction for children with hydrocephalus presenting to the emergency department (ED).

To evaluate whether a change in the optic nerve sheath diameter (ONSD) from prior asymptomatic baseline was associated with shunt failure.

Participants in this prospective single-center observational cohort study at a tertiary care children’s hospital were enrolled from January 5, 2018, to March 2, 2022. Children 18 years or younger with cerebrospinal fluid (CSF) shunts were consecutively recruited during routine asymptomatic visits in the outpatient neurosurgery clinic. The institution sees approximately 1000 children with shunts. Children with comorbid eye pathologic conditions known to increase the ONSD were excluded, as were those with shunt interventions between baseline and symptomatic ocular POCUS. Data analysis was completed in May 2024.

Baseline ocular POCUS was performed on asymptomatic children with CSF shunts recruited in the outpatient neurosurgery clinic; a second ocular POCUS was performed if the patient subsequently presented to the ED with symptoms of shunt failure. Change in ONSD from prior asymptomatic baseline to symptomatic presentation was the main study exposure.

Shunt failure was defined as intraoperative confirmation of inadequate CSF flow through the shunt system associated with identifiable shunt complications, including catheter or valve obstruction, shunt tubing fracture or disconnection, or proximal catheter migration out of the ventricle within 96 hours from presentation to the ED.

Seventy-six pairs of baseline and symptomatic ultrasonograms from 58 patients (mean [SD] age, 6.6 [4.7] years; 36 of 58 boys [62%]) were included. Twenty patients (35%) had 2 or more prior shunt revisions, and 29 (50%) had communicating hydrocephalus. The primary outcome of intraoperatively confirmed shunt failure was observed in 14 of 76 ED patient presentations (18%). The mean (SD) change in ONSD among patients with shunt failure was 0.89 (0.66) mm vs 0.16 (0.40) mm among patients without shunt failure (mean difference, 0.73 mm [95% CI, 0.47-0.99 mm]; P < .001). The odds of full shunt failure were 1.4 times higher (95% CI, 1.21-1.78; P < .001) for every 0.1-mm increase in ONSD. The area under the receiver operating characteristic curve was 0.86, with an optimal cutoff of 0.4 mm or more, yielding a sensitivity of 0.93, specificity of 0.73, positive predictive value of 0.43, and negative predictive value of 0.98, for a disease prevalence of 15%, with a positive likelihood ratio of 3.39 and negative likelihood ratio of 0.10.

In this single-center, prospective cohort study of symptomatic children with CSF shunts, a change in ONSD measured by ocular POCUS was associated with shunt failure. A change in ONSD from prior asymptomatic baseline to symptomatic presentation of less than 0.4 mm identified a population at low risk of true shunt failure. Further research is warranted to validate these findings, the presented change in ONSD threshold, and to risk stratify low-risk patients for cross-sectional neuroimaging.

This cohort study evaluates whether a change in optic nerve sheath diameter from prior asymptomatic baseline was associated with shunt failure among symptomatic children with cerebrospinal fluid shunts.