Correlation of Optic Nerve Sheath Diameter Measured Using Bedside Ultrasonography with Directly Measured Intracranial Pressure

INTRODUCTION: Monitoring and managing intracranial pressure (ICP) is a cornerstone of treatment in neurocritical care, particularly in patients with traumatic brain injury (TBI). While the gold standard for ICP monitoring is an external ventricular drain (EVD), this invasive procedure carries risks such as hematoma formation and infections. Furthermore, EVD placement may not be feasible in all critical care settings due to the need for specialized nursing care. Consequently, there is growing interest in noninvasive methods for ICP assessment. In this study, we investigated the correlation between intracranial pressure and optic nerve sheath diameter (ONSD) measured using bedside ultrasound, a simple and noninvasive technique. METHODS: In 30 adult patients, who were sedated and mechanically ventilated, 131 paired ICP and ONSD readings were recorded daily until EVD removal or extubation. ICP was transduced to a monitor, and ONSD was measured via ultrasound. Since the data were not normally distributed, the Spearman Rank Correlation was used to assess the relationship between ICP and ONSD. Mixed-effects models accounted for repeated measures and patient variability, with ICP as the dependent variable, ONSD (average or higher) as the fixed effect, and patient ID as the random effect. A Receiver Operating Characteristic (ROC) curve determined an ONSD cut-off for predicting raised ICP (>20 mmHg). RESULTS: The Spearman correlation coefficient was 0.144 (p = 0.1007), indicating a weak, non-significant positive relationship between ONSD and ICP. The mixed-effects model estimated that each 1 mmHg increase in ICP corresponds to a 0.008 mm increase in ONSD (p = 0.101). The ROC curve showed poor predictive ability (AUC = 0.574), with an ONSD cut-off of 5.98 mm yielding 55.6% sensitivity and 71.2% specificity.(1–6) DISCUSSION AND CONCLUSION: The correlation of Optic Nerve sheath Diameter with ICP has been studied many times before. Though in contrast to our study, earlier studies have found a stronger correlation between the two. However, each study has given a different threshold from their ROC curves. The above table shows the variability in the ONSD cut-off given by each study to determine raised ICP. The inconsistency of cut-offs across studies and the lack of statistically significant findings in our dataset suggest that ONSD on its own cannot reliably predict intracranial pressure. Overall, while ONSD may offer some predictive value for ICP, our study reinforces the need for caution in using it as a standalone diagnostic tool.