Background: Acute obstructive hydrocephalus secondary to spontaneous intracerebral/intraventricular hemorrhage (ICH/IVH) requires early cerebrospinal fluid (CSF) drainage to reduce intracranial pressure (ICP). Extensive CSF drainage may reduce IVH clot burden. We characterize CSF dynamics, strategies and impact on end of treatment (EOT) IVH volume (72 hours post randomization [Rand]) in the CLEAR III trial.
Methods: Prospective analysis of CSF output in all 500 patients enrolled in the CLEAR III trial, a multicenter, double-blind, randomized study comparing EVD + intraventricular recombinant tissue plasminogen activator (rtPA) vs. EVD + placebo for treatment of obstructive IVH and intracerebral hemorrhage (ICH) volume <30cc. CSF output was recorded every 4 hours until 7 days post Rand, and compared by clinical and radiological variables.
Results: Daily median CSF output in the first week was 188cc (IQR: 125, 252). Maximum daily EVD drip settings were <10mmHg in 27.8%, =10 in 44.1% and >10 in 28.1%. Independent predictors of higher daily CSF output after adjustment for initial IVH volume (p=0.04) were lower drip setting (p<0.001), lower age (p<0.001), male sex (p=0.03), dual EVD (p=0.005), CSF protein (p<0.001) and ICP>20mmHg (P=0.007). Both EOT IVH volume and change in IVH volume (ChgIVH) (over 1st week) were independently associated with total CSF output (P=0.004/<0.001, respectively), and initial IVH volume (P<0.001/<0.001)). Early opening of 3rd and 4th ventricle (P=0.03) was associated with lower EOT volumes, while CSF protein (P=0.02), and side of EVD ipsilateral to largest IVH (P=0.04) were associated with ChgIVH. Shunting for hydrocephalus was performed in 18.6% over 1 year follow-up and was associated with higher total CSF output over first week (P<0.001).
Conclusions: CSF circulation in severe IVH can be assessed by CSF output which is associated with EVD drip management and other clinical variables. EOT IVH volume and IVH volume reduction are important surrogate endpoints which are related to CSF dynamics. VP shunt requirement in spontaneous IVH is associated with early CSF output levels. These results permit future correlation of CSF output with treatment rendered (thrombolysis versus placebo) with upcoming unblinding of the trial.