Evidence Library

Current evidence framework · versioned by data year, population and definition

This library explains which findings support the next research steps at Ethical Saving: AIRCHILL mechanism data, preclinical models, external clinical signals, ongoing randomised trials, epidemiology and health-economic anchors. External animal and human studies are design knowledge and scientific justification; they are not presented as proven AIRCHILL patient efficacy.

Project evidence

• BMBF funding: €499,999 for two years of “Airchill – Gekühlte Beatmung” at UKE is publicly documented; the website rounds this to €500k.
• AIRCHILL programme: documented overall programme includes 14 pigs and 11 healthy volunteers. Published subsets are smaller.
• Patient use: AIRCHILL has treated 0 patients.
• Patents: granted EP3509683B1 and US11395900B2 are linked through public patent records.

Versioned German epidemiology

• OHCA: about 136,000 out-of-hospital cardiac arrests annually as the broad event denominator; this is not the same as EMS resuscitation attempts.
• EMS resuscitation attempts: roughly 54,000–67,000 per year depending on the DGRR denominator and extrapolation.
• Temperature management: 17.3% of hospital-admitted OHCA patients in the 2024 DGRR reference group; this is a care rate, not a treatment target.
• Mechanical thrombectomy: 22,445 procedures in a 2023 peer-reviewed DRG series and 21,609 in a 2024 national claims analysis with different methodology.
• Neonatal HIE: historical German observation around 543 cases/year; a modelled figure around 981/year is derived from 2025 births × an international 1.5/1,000 incidence anchor and is not a measured German incidence.
• Moderate/severe TBI: historical German trauma-registry data provide a national scale anchor; current AIRCHILL eligibility remains a separate data gap.

AIRCHILL mechanism & preclinical rationale

Sedlacik et al. demonstrated measurable regional brain cooling in pigs using high-flow cold air and MRI temperature mapping. Animal studies by other groups in cardiac arrest and focal ischaemia support the broader hypothesis that timing can be an important part of the thermal dose. These studies justify testing timing, route, dose and safety; they do not prove human AIRCHILL efficacy.

Cardiac arrest

Relevant evidence includes PRINCE, PRINCESS, the pooled PRINCE/PRINCESS analysis, PRINCESS2, TTM, HYPERION and TTM2. PRINCESS was neutral in the overall primary endpoint but timing and rhythm signals inform more precise future study design. Current ERC/ESICM guidance emphasises continuous temperature control and fever prevention rather than routine deep hypothermia for every comatose post-ROSC patient.

Stroke

Relevant programmes include COTTIS, the randomised COTTIS-2 follow-up, CHILL-ART and COOLHEAD-2b. A small COTTIS matched-pair analysis reported a striking functional-outcome signal, but its size and non-randomised design require randomised confirmation. Ethical Saving therefore uses much smaller +1/+3/+5 percentage-point scenarios as sensitivity analyses rather than extrapolating that observed difference.

Neonatal HIE

Therapeutic hypothermia is established for selected moderate-to-severe neonatal HIE. This provides biological precedent for cooling but does not prove incremental benefit from a new transport or early-cooling device. Any paediatric AIRCHILL pathway would require a separate safety, dosing, development and regulatory programme.

Traumatic brain injury

POLAR, Eurotherm3235 and LTH-1 show why phenotype, intracranial pressure, timing, duration, rewarming and safety matter. Broad prophylactic hypothermia has not established a universal outcome benefit; selected subgroup signals remain useful design information.

Concussion

External head/neck cooling studies suggest a separate research opportunity, mainly around symptom recovery and return to activity. Typical concussion patients are awake and do not require invasive ventilation, so this is not a current respiratory AIRCHILL use case.

Heatstroke

Rapid cooling is well supported, particularly for exertional heatstroke. Cooling rate and time to safe core temperature are stronger initial feasibility endpoints than an assumed survival benefit. Respiratory cooling would have to be studied as a complementary option where established rapid-cooling methods are unavailable or insufficient.

Ongoing trials

The wider research landscape includes PRINCESS2, ICECAP, P-ICECAP, COTTIS-2 and other registered temperature-management studies. Registration means a study is planned or ongoing; it does not establish benefit.

Health economics

External cost studies show strong gradients by neurological disability after stroke, TBI and hypoxic brain injury. These are useful anchors for modelling but are not German reimbursement values and do not justify a guaranteed national savings claim.

Sensitivity scenarios

+1, +3 and +5 percentage-point scenarios are explicit planning sensitivities, not AIRCHILL effect estimates. They are chosen to test whether even small absolute effects could be clinically and economically relevant and to inform future sample-size planning once an AIRCHILL feasibility/pilot programme produces its own measured effect estimate.

Evidence-status rule

Where data are transferred from another country, mathematically derived, operational assumptions or still missing, they should remain labelled TRANSFERRED, DERIVED, ASSUMPTION or DATA GAP. Promising evidence is framed positively as a reason for controlled study while keeping the boundary between scientific rationale and proven clinical benefit explicit.

Full hypothermia research register → · Sources, assumptions & gaps →