Efficacy and safety of combined intraventricular fibrinolysis with lumbar drainage for prevention of permanent shunt dependency after intracerebral hemorrhage with severe ventricular involvement: A randomized trial and individual patient data meta-analysis.

OBJECTIVE: Intraventricular hemorrhage (IVH) is a negative prognostic factor in intracerebral hemorrhage (ICH) and is associated with permanent shunt dependency in a substantial proportion of patients post-ICH. IVH treatment by intraventricular fibrinolysis (IVF) was recently linked to reduced mortality rates in the CLEAR III study and IVF represents a safe and effective strategy to hasten clot resolution that may reduce shunt rates. Additionally, promising results from observational studies reported reductions in shunt dependency for a combined treatment approach of IVF plus lumbar drains (LDs). The present randomized, controlled trial investigated efficacy and safety of a combined strategy-IVF plus LD versus IVF alone-on shunt dependency in patients with ICH and severe IVH. METHODS: This randomized, open-label, parallel-group study included patients aged 18 to 85 years, prehospital modified Rankin Scale ≤3, ICH volume < 60ml, Glasgow Coma Scale of <9, and severe IVH with tamponade of the third and fourth ventricles requiring placement of external ventricular drainage (EVD). Over a 3-year recruitment period, patients were allocated to either standard treatment (control group receiving IVF consisting of 1mg of recombinant human tissue plasminogen activator every 8 hours until clot clearance of third and fourth ventricles) or a combined treatment approach of IVF and-upon clot clearance of third and fourth ventricles-subsequent placement of an LD for drainage of cerebrospinal fluid (CSF; intervention group). The primary endpoint consisted of permanent shunt placement indicated after a total of three unsuccessful EVD clamping attempts or need for CSF drainage longer than 14 days in both groups. Secondary endpoints included IVF- and LD-related safety, such as bleeding or infections, and functional outcome at 90 and 180 days. Conducted endpoint analyses used individual patient data meta-analyses. The study was registered at clinicaltrials.gov (NCT01041950). RESULTS: The trial was stopped upon predefined interim analysis after 30 patients because of significant efficacy of tested intervention. The primary endpoint was analyzed without dropouts and was reached in 43% (7 of 16) of the control group versus 0% (0 of 14) of the intervention group (p = 0.007). Meta-analyses were based on overall 97 patients, 45 patients receiving IVF plus LD versus 42 with IVF only. Meta-analyses on shunt dependency showed an absolute risk reduction of 24% for the intervention (LD, 2.2% [1 of 45] vs no-LD, 26.2% [11 of 42]; odds ratio [OR] = 0.062; confidence interval [CI], 0.011-0.361; p = 0.002). Secondary endpoints did not show significant differences for CSF infections (OR = 0.869;CI, 0.445-1.695; p = 0.680) and functional outcome at 90 days (OR = 0.478; CI, 0.190-1.201; p = 0.116), yet bleeding complications were significantly reduced in favor of the intervention (OR = 0.401; CI, 0.302-0.532; p < 0.001). INTERPRETATION: The present trial and individual patient data meta-analyses provide evidence that, in patients with severe IVH, as compared to IVF alone, a combined approach of IVF plus LD treatment is feasible and safe and significantly reduces rates of permanent shunt dependency for aresorptive hydrocephalus post-ICH. ANN NEUROL 2017;81:93-103.