A comparison of non-invasive versus invasive measures of intracranial pressure in hypoxic ischaemic brain injury after cardiac arrest

Aim: Increased intracranial pressure (ICP) in hypoxic ischaemic brain injury (HIBI) can cause secondary ischaemic brain injury and culminate in brain death. Invasive ICP monitoring is limited by associated risks in HIBI patients. We sought to evaluate the agreement between invasive ICP measurements and non-invasive estimators of ICP (nICP) in HIBI patients. Methods: Eligible consecutive adult (age >18) cardiac arrest patients with HIBI were included as part of a single centre prospective interventional study. Invasive ICP monitoring and nICP measurements were undertaken using: a) transcranial Doppler ultrasonography (TCD), b) optic nerve sheet diameter ultrasound (ONSD) and c) jugular venous bulb pressure (JVP). Multiple measurements applied in linear mixed-effects models were considered to obtain the correlation coefficient between ICP and nICP as well as their predictive abilities to detect intracranial hypertension (ICP >= 20 mm Hg). Results: Eleven patients were included (median age of 47 [range 20-71], 8 males and 3 females). There was a linear relationship between ICP and nICP with ONSD (R=0.53 [p 0.0001]), JVP (R = 0.38 [p 0.001]) and TCD =0.30 [p 0.01]). The ability to predict intracranial hypertension was highest for ONSD and TCD (area under the receiver operating curve (AUC)= 0.96 [95% CI: 0.90-1.00] and AUC =0.91 [95% CI: 0.83-1.00], respectively). JVP presented the weakest prediction ability (AUG = 0.75 [95% CI: 0.56-0.94]). Conclusions: ONSD and TCD methods demonstrated agreement with invasively-monitored ICP, suggesting their potential roles in the detection of intracranial hypertension in HIBI after cardiac arrest.