Respiratory Trials·org

Topic

Neurology in critical care

Cardiac arrest, traumatic brain injury, stroke, and neuroprognostication.

Stroke thrombolysis has moved from the clock to the tissue. It began with a three-hour window and an explicit trade — better function, more intracranial bleeding (NINDS t-PA) — extended to 4.5 hours (ECASS III). Imaging then dissolved the window: MRI mismatch identified wake-up strokes that still benefited (WAKE-UP), perfusion imaging extended it to nine hours (EXTEND), and then to twenty-four (HOPE). For large-vessel occlusion, thrombectomy works up to 24 hours with the right mismatch (DAWN), and bridging thrombolysis before it should be retained (Tenecteplase before thrombectomy).

After cardiac arrest, cooling to 33°C was no better than actively maintained normothermia (TTM2) — fever prevention survived, deliberate hypothermia did not.

Transfusion thresholds in brain injury are contested. A liberal threshold improved neurological outcome in acute brain injury generally (TRAIN) but not in traumatic brain injury (HEMOTION) or subarachnoid haemorrhage (SAHaRA). Two of three are negative, which is where the weight of evidence sits.

In intracerebral haemorrhage, driving systolic pressure to 110-139 mmHg gave no benefit over a 140-179 target (ATACH-2). After subarachnoid haemorrhage, nimodipine remains the only drug that improves outcome, on the strength of a trial from 1989 (BRANT).

Thrombolysis beyond the clock

Trial Year Therapy Effect Finding
HOPE 2025 Alteplase Imaging-selected thrombolysis worked as late as 24 hours from onset
EXTEND 2019 Alteplase guided by perfusion imaging Perfusion imaging identified patients who benefited from thrombolysis up to 9 hours out
WAKE-UP 2018 Alteplase guided by MRI mismatch MRI mismatch identified wake-up strokes that still benefited from thrombolysis

Transfusion in brain injury

Trial Year Therapy Effect Finding
SAHaRA 2025 Liberal transfusion strategy Liberal transfusion did not improve outcomes after subarachnoid haemorrhage
HEMOTION 2024 Liberal transfusion strategy A liberal transfusion threshold did not improve outcomes in traumatic brain injury
TRAIN 2024 Liberal transfusion strategy A liberal transfusion threshold improved neurological outcome in acute brain injury

Thrombolysis for ischaemic stroke

Trial Year Therapy Effect Finding
ECASS III 2008 Alteplase Alteplase still worked between 3 and 4.5 hours, with the same bleeding cost
NINDS t-PA 1995 Alteplase Thrombolysis within 3 hours improved outcome despite causing more intracranial bleeding

Blood pressure in intracerebral haemorrhage

Trial Year Therapy Effect Finding
ATACH-2 2016 Intensive blood pressure lowering Aggressive blood pressure lowering in intracerebral haemorrhage gave no benefit

Bridging thrombolysis before thrombectomy

Trial Year Therapy Effect Finding
Tenecteplase before thrombectomy 2025 Tenecteplase Giving tenecteplase before thrombectomy beat thrombectomy alone

Extubating the brain-injured patient

Trial Year Therapy Effect Finding
Coplin 2000 prospective cohort 2000 Prompt extubation Delaying extubation in brain-injured patients who were ready nearly doubled pneumonia and doubled ICU stay

Late thrombectomy

Trial Year Therapy Effect Finding
DAWN 2018 Mechanical thrombectomy Thrombectomy worked up to 24 hours out when imaging showed salvageable brain

Preventing delayed ischaemia after subarachnoid haemorrhage

Trial Year Therapy Effect Finding
BRANT 1989 Nimodipine Oral nimodipine reduced cerebral infarction and improved outcome after subarachnoid haemorrhage

Prognostication after cardiac arrest

Trial Year Therapy Effect Finding
Coppler 2016 prognostic cohort 2016 Objective futility criteria Three objective criteria identified cardiac arrest patients with essentially no chance of survival

Temperature after cardiac arrest

Trial Year Therapy Effect Finding
TTM2 2021 Targeted hypothermia Cooling to 33°C after cardiac arrest was no better than simply preventing fever

Key references

Guidelines, standards and reviews for this topic.