Noninvasive Intracranial Pressure Estimation Using Transcranial Doppler to Rule Out Intracranial Hypertension in Children Undergoing Lumbar Puncture: A Feasibility Study.

Background

Intracranial pressure (ICP) monitoring is essential in pediatric neurocritical care, but current gold-standard methods are invasive, associated with relevant risks and necessity of sedation. Ultrasonographic methods, such as transcranial Doppler (TCD), would allow a noninvasive ICP estimation, permitting repeated samplings and serial monitoring. This study aimed to evaluate the feasibility and accuracy of TCD-derived ICP compared with cerebrospinal fluid opening pressure in children undergoing lumbar puncture.

Methods

Prospective observational study at the University Hospital of Padua, Italy, including pediatric patients requiring diagnostic lumbar puncture. ICP was assessed invasively, measuring cerebrospinal fluid opening pressure and simultaneously estimated noninvasively using TCD-derived cerebral hemodynamic parameters: (i) pulsatility index derived methods (Bellner and Bellner-modified equation) and (ii) Czosnyka-equation. Accuracy and predictive capability of the best test were then tested.

Results

We included 60 children (median age 7.43 years). Among the tested methods, Czosnyka equation showed the best agreement with invasive ICP with mean difference of 3.78 mmHg [5.1 cmH2O] and fair but significant correlation (Spearman rho = 0.29, P = 0.027). Accuracy was calculated for a threshold of 20 mmHg [27.2 cmH2O] resulting in sensitivity 28.6%, specificity 91.3%, positive predictive value 50% and negative predictive value 80.8%, suggesting a good ability to exclude intracranial hypertension. Pulsatility index-based methods showed poor agreement.

Conclusions

TCD-derived ICP estimation using Czosnyka equation may represent a useful noninvasive adjunct for ruling out intracranial hypertension in children. However, the limited sensitivity and accuracy preclude its use as a standalone diagnostic tool. Larger studies are needed to validate its role within multimodal monitoring strategies.