Cold-water immersion during out-of-hospital heatstroke cardiac arrest resuscitation: feasible but futile? Authors: Clément Derkenne, Lucas Boussarie, Miguel Peireira, Nadia Nakhil, Matthieu Heidet, Paris Fire Brigade Cardiac Arrest Task Force Published: 2026 Dear Editor, Since 17 June 2026, France has faced a severe heatwave, with temperatures reaching 45°C. Across the greater Paris area, the Paris Fire Brigade has deployed an aggressive cooling strategy: within 15 min, firefighters immerse heatstroke patients in a portable bathtub at 14°C with 60 kg of added ice while a physician-staffed Mobile Intensive Care Unit (MICU) manages life-threatening complications before and/or during, and/or after immersion. This protocol has been applied over 800 times within six days; quantifying how many immersions coincided with cardiac arrest before, during, or after cooling remains challenging amid a crisis of this scale. We report two cases of out-of-hospital cardiac arrest (OHCA) resuscitated during cold-water immersion. Both patients were comatose at Basic Life Support teams arrival (core temperatures of 42.4°C and 43°C) and progressed to arrest with persistent asystole. Advanced life support – mechanical chest compressions (LUCAS 2, Stryker), mechanical ventilation (Monnal T60, Air Liquide), and intraosseous epinephrine – were delivered while immersed, with electrical devices kept out of the water (Fig. 1). Mobile extracorporeal cardiopulmonary resuscitation (ECPR) teams did not identify any indication for cannulation, given persistent asystole, absence of any sign of life, and the historically poor prognosis of hyperthermic OHCA. Resuscitation was discontinued after 52 and 30 min, respectively, with slower-than-expected cooling (−0.08 and −0.07°C/min).Fig. 1Advanced life support delivered during cold-water immersion drawing and rectal temperature curve. The patient's entire body is immersed with the exception of the airways, which are in any case protected by orotracheal intubation. All electrically powered devices – the mechanical chest-compression device, the ventilator, and the monitor/defibrillator – are kept out of the water at all times. The patient's arms remain outside the immersion bath in order to secure the chest-compression device's straps, reducing the immersed surface area relative to non-arrest immersion protocols. The devices used carry manufacturer-rated splash protection (LUCAS 2/3: IP43; Monnal T60: IP34), corresponding to protection against spraying or splashing water rather than submersion, consistent with their positioning at the edge of the immersion tub rather than within the bath itself. Advanced life support delivered during cold-water immersion drawing and rectal temperature curve. The patient's entire body is immersed with the exception of the airways, which are in any case protected by orotracheal intubation. All electrically powered devices – the mechanical chest-compression device, the ventilator, and the monitor/defibrillator – are kept out of the water at all times. The patient's arms remain outside the immersion bath in order to secure the chest-compression device's straps, reducing the immersed surface area relative to non-arrest immersion protocols. The devices used carry manufacturer-rated splash protection (LUCAS 2/3: IP43; Monnal T60: IP34), corresponding to protection against spraying or splashing water rather than submersion, consistent with their positioning at the edge of the immersion tub rather than within the bath itself. The 2025 ERC guidelines recommend cold-water immersion for heatstroke without arrest, and the universal ALS algorithm once arrest occurs.1 We are unaware of any previous ALS report during active immersion; the only related report describes synchronized cardioversion of a heatstroke patient with ventricular tachycardia, performed without removing them from the tub.2 These cases confirm mechanical compression, ventilation, and drug administration are feasible during immersion, provided electrical equipment stays out of the water. The reduced cooling efficacy likely reflects the circulatory physiology of cardiac arrest: heat transfer from core to periphery is flow-dependent, and mechanical compressions generate cardiac output well below physiological levels, limiting heat dissipation regardless of water temperature. Reduced water agitation, precluded for electrical safety, and reduced immersed surface, since the arms remained outside to secure the device, likely contributed further. By comparison, large series of non-arrest immersion report cooling rates of 0.2–0.3°C/min.3 Decision-making around ECPR eligibility in hyperthermic OHCA is hampered by the absence of a prognostic tool comparable to the HOPE score in accidental hypothermia.4 Literature is limited to case reports, including venous-arterial-ECPR before arrest in refractory heatstroke shock.5 Beyond these findings, two observations have shaped our approach. First, OHCA mortality has been strikingly high, likely reflecting prolonged no-flow linked to social isolation and the known link between hyperthermic OHCA to poor survival. Second, despite a five- to tenfold rise in OHCA incidence, no collective decision to withhold resuscitation has been made: every patient receives individualized assessment and MICU response wherever any prospect of survival remains. This report is limited by its size and differing temperature sites; larger prospective series are needed to confirm whether ALS during immersion is genuinely ineffective in cardiac arrest. CRediT authorship contribution statement Clément Derkenne: Writing – original draft, Supervision, Investigation, Data curation, Conceptualization. Lucas Boussarie: Investigation, Data curation, Conceptualization. Miguel Peireira: Methodology, Data curation, Conceptualization. Nadia Nakhil: Investigation, Data curation, Conceptualization. Matthieu Heidet: Writing – review & editing, Investigation, Formal analysis, Conceptualization. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.