Accidental hypothermia in the Southeastern United States: the role of extracorporeal membrane oxygenation and the need for regional protocol development—a case series

Introduction

Hypothermia, defined as T_core <35 ℃, denotes a critical transition at which body heat loss exceeds heat generation (1). In the United States (US), accidental hypothermia causes roughly 1,500 deaths annually. The incidence of accidental hypothermia and other cold-related morbidity in the warmer climates of the Southern US is notably lower, accounting for <17% of all US cases (2). The gold standard for core body temperature assessment is placement of a temperature probe in the distal third of the esophagus, as this closely estimates the temperature of the myocardium if placed correctly (3).

For appropriate prehospital triage, the Swiss staging system for hypothermia encompasses a spectrum of core temperatures and clinical appearances that can be utilized to assess and triage hypothermia (4-6). Stage I corresponds to an individual with T_core 35–32 ℃. Compensatory shivering often accompanies mild. Management includes passive techniques to abate exposure, including introducing a warm environment, clothing, and oral fluids. Stage II presents with an alteration in consciousness with or without compensatory shivering, and a T_core <32–28 ℃. Active external rewarming techniques, including heat packs, warm parenteral fluids, and full-body insulation, are appropriate treatments at this stage. The patient should also have continuous cardiac and T_core monitoring. Stage III: the individual is unconscious, has vital signs, and T_core <28 ℃. Treatment for Stage III includes Stage II treatments, airway management, and transfer to extracorporeal membrane oxygenation (ECMO)-capable centers. Stage IV defines a hypothermic patient with absent vital signs who requires cardiopulmonary resuscitation (CPR) (4-6).

The revised Swiss Hypothermia Score (RSHS) provides a more feasible prehospital staging of accidental hypothermia without the need for a core temperature, a metric that can be difficult to accurately obtain in the field. This pattern is demonstrated in the cases below with discrepancy between prehospital and Emergency Department (ED) initial temperatures. Instead, the RSHS entails a purely qualitative clinical assessment of responsiveness to stimuli, known as AVPU (alert, verbal, pain, unconscious), which has been demonstrated to correlate with core temperature linearly, and thus serves as an appropriate surrogate (5).

The following five cases of severe accidental hypothermia entail presentations to a Southeastern US level one trauma center with ECMO capability between 2015 and 2024. The following patients presented with T_core <28 ℃ in the setting of known environmental exposure and hemodynamic instability. They were all subsequently vascularly cannulated and placed on venovenous (VV) or venoarterial (VA) ECMO. A review of prehospital and hospital documentation was utilized to retrospectively follow patient care until death or hospital discharge. Neuroprognostication of each case was determined through utilization of a cerebral performance category (CPC) score. We present this article in accordance with the AME Case Series reporting checklist (available at https://jeccm.amegroups.com/article/view/10.21037/jeccm-2025-35/rc).

Case presentation

Case 1

A 47-year-old man with a past medical history of diabetes, alcohol use disorder, and seizures was found down in his driveway in March by emergency medical services (EMS) for an unknown duration of time. He was pulseless and hypothermic with a temperature of 20.5 ℃ when EMS arrived. In the prehospital setting, after two minutes of manual CPR, return of spontaneous circulation (ROSC) was achieved following 1 mg of epinephrine, 2 mg of naloxone, 20 units of vasopressin, and a single defibrillation. CPR was initiated after the loss of palpable pulses during transport. ROSC occurred after one minute. The patient was transferred to the ED while receiving warmed parenteral fluids and external support from warming blankets.

He presented to the ED with fixed and dilated pupils with an initial T_core of 23.1 ℃ rectally. This was much warmer than the initial EMS temperature, posing further questions about the accuracy of temperatures taken in the prehospital setting. Head imaging showed no acute abnormality. During active rewarming in the ED, he had an additional cardiac arrest. He was intubated and received six minutes of manual CPR, including 1 mg of epinephrine and one defibrillation, with ROSC achieved. Following this event, the patient awoke and was following commands. The decision was made to proceed with venous cannulation for continued aggressive, active rewarming using VV-ECMO given refractory acidosis without the need for mechanical cardiac support. After two hours of active internal rewarming, he was still in the ED, and his temperature had improved to 33.3 ℃. He was able to be decannulated from ECMO and extubated in the ED. The patient was subsequently transferred to the intensive care unit (ICU) for further medical management off vasopressors and mechanical ventilation.

The patient left the hospital against medical advice four days later despite recommendations for automatic implantable cardioverter-defibrillator (AICD) placement. When the patient left the hospital, he had a CPC score of 1. He subsequently presented 20 months later to the ED in cardiac arrest and was found to be hypothermic again with a T_core of 27.4 ℃. At this time, the decision was made not to cannulate for ECMO in the setting of a known active variceal bleeding, a contraindication for ECMO-related anticoagulation. He was admitted for continued medical management and died the following day after being transitioned to comfort care measures.

Case 2

A 58-year-old female with a past medical history of heart failure, pulmonary hypertension, peripheral arterial disease, and substance use disorder was found minimally responsive on a chair inside a barn without heat in January. Upon EMS arrival, the patient was found bradycardic with a heart rate in the 20s with no appreciable manual blood pressure. She was transported without CPR.

Upon arrival at the ED, the patient was pulseless and hypothermic with a T_core of 26.5 ℃ rectally. She received CPR for nine minutes, and ROSC was achieved. Active rewarming efforts, including warm parenteral fluids and external rewarming devices, were initiated. However, she had a second cardiac arrest. The ECMO team was consulted for ongoing cardiopulmonary support in the setting of accidental hypothermia. CPR was continued for an additional 29 minutes. During these codes, she received a total of 4 g of calcium chloride, ten doses of 1 mg of epinephrine, 2 mg of naloxone in a single dose, and three doses of 50 mEq sodium bicarbonate. She was successfully placed on VA-ECMO in the ED and was transferred to the ICU. The patient was later found to have outside hospital documentation for a do-not-resuscitate order (DNR). Following a discussion with the family, the patient was transitioned to comfort care measures and died the following day. When care was withdrawn, the patient had a CPC score of 5.

Case 3

A 23-year-old male with a history of intravenous drug use was discovered minimally responsive in a backyard in March for an unknown duration by EMS. He was initially combative with EMS, but palpable pulses were not appreciated during transport to the ED. In the prehospital setting, he received four minutes of manual CPR, including one defibrillation and one dose of epinephrine. Subsequently, ROSC was achieved.

A supraglottic airway device was in place upon arrival at the ED, and the patient was found to be bradycardic and hypothermic with an initial temperature of 27.2 ℃. He had several episodes of cardiac arrest in the ED, intermittently going into ventricular fibrillation requiring chest compressions and defibrillation nine times. He was also placed on norepinephrine and amiodarone drips, which were continued during the arrests. A perfusing rhythm was achieved prior to vascular cannulation, and the patient was successfully placed on VV-ECMO in the ED. He was requiring vasopressor support with norepinephrine at that time and was subsequently placed on dobutamine in the ICU. He was found to have a T_core of 21.5 ℃ and severe metabolic acidosis at the time ECMO was initiated. The supraglottic airway was exchanged for an endotracheal tube after ECMO was started and the patient was taken to the ICU. Eight days later, he had a complete recovery of his ejection fraction and was discharged home from the hospital. At the time of discharge, he had a CPC score of 1.

Case 4

A 56-year-old female with a past medical history of substance use disorder and hypertension was discovered in January on the floor of an apartment without heat. She was found to have slow respirations and a weak pulse. In the prehospital setting, she was given naloxone by the fire department with no response and subsequently lost pulses when EMS arrived at the scene. She received 69 minutes of prehospital CPR, including 36 minutes of manual and 33 minutes of device-assisted compressions. During this time, she was defibrillated 21 times, given 2 mg of naloxone once, 1 mg of epinephrine, and an initial amiodarone dose of 300 mg followed by an additional 150 mg of amiodarone, and 2 g of magnesium.

Upon presentation to the ED, the patient remained in persistent ventricular fibrillation with continued device-assisted compressions for 44 additional minutes, with alternating ventricular fibrillation and pulseless electrical activity (PEA). During this time, she received 4 mg of naloxone, one amp of calcium chloride, 2 mg of epinephrine, 35 mg of Esmolol IV push, and 20 units of vasopressin. While she received warmed parenteral fluids with a forced-air external rewarming device, she was successfully cannulated for VA-ECMO and transferred to the ICU. While in the ICU, she had a recovery of her systolic cardiac function and was decannulated seven days after initiation of ECMO. Subsequently, the patient became oliguric, necessitating continuous renal replacement therapy (CRRT) for 9 days. She remained intubated and with minimal purposeful neurological activity during this period. She was able to be extubated and required oxygen by a high-flow nasal cannula thereafter. Three days following extubation, she still had a minimal return of purposeful neurological functioning, with a CPC score of 3, and the decision was made by the family to transition to comfort care, and she subsequently died 22 days after her initial presentation to the ED.

Case 5

A 64-year-old female with a past medical history of schizophrenia and cocaine use was found unresponsive at a bus stop in November for an unknown duration in asystole arrest. EMS initiated manual CPR for 10 minutes, with an additional 12 minutes of mechanical device-assisted compressions for a total of 22 minutes. She received two 1 mg doses of epinephrine and a single defibrillation by EMS, before hospital arrival.

She presented to the ED in asystole and hypothermic with a T_core of 20.0 ℃, the supraglottic airway in place, and ongoing CPR. She was noted to have fixed pupils and no obvious signs of trauma or hemorrhage. In the ED, the patient underwent 20 minutes of CPR with four doses of 1 mg of epinephrine followed by an epinephrine drip and 50 mEq of sodium bicarbonate. An external warming device was applied, and approximately 2 and 3 L of warmed-parenteral fluids were given. During CPR, she exhibited one shockable rhythm with subsequent defibrillation, followed by PEA. ROSC was achieved. She was subsequently placed on VA-ECMO post-ROSC and transferred to the ICU. She was decannulated from ECMO 3 days later, at which time she was able to say a few words and follow commands with all extremities. She had a prolonged hospital stay complicated by the development of seizures necessitating multiple antiepileptic drugs (AEDs), a hypertensive emergency requiring a nicardipine drip, and bacteremia with methicillin-resistant staphylococcus aureus (MRSA) and vancomycin-resistant enterococcus (VRE). She was able to be titrated to oral medications for her hypertension, had a resolution in her bacteremia, and after a 41-day hospitalization with multiple pathologies, she was ultimately discharged to a skilled nursing facility with a CPC score of 3.

Ethical approval

All procedures performed in this study were in accordance with the ethical standards of the institutional and/or national research committee(s) and with the Declaration of Helsinki and its subsequent amendments. The publication of this case series was waived from patients’ consent, according to the Advocate Health – Wake Forest University School of Medicine Institutional Review Board.

Discussion

ECMO offers life-saving utility in the setting of Stage III or Stage IV with hemodynamic instability [life-threatening dysrhythmia, hypotension with systolic blood pressure (SBP) <90 mmHg, respiratory failure, and refractory acidosis] (4-6). Early recognition of appropriate patients and prompt transport to ECMO-capable medical centers results in drastic mortality benefit, with an absolute risk reduction of 42% (7).

This is further supported by the ICE-CRASH trial, which demonstrated improved 28-day survival and favorable neurologic outcomes in hypothermic patients appropriately cannulated for ECMO compared to those who did not receive vascular cannulation. Improvements in 28-day survival, neurologic outcomes, and event-free days, which were described as ICU-, ventilator-, renal replacement therapy (RRT)-, and catecholamine administration-free days, were also demonstrated in appropriately selected hypothermic patients who did not present with cardiac arrest (8).

VV-ECMO can be considered in a subset of Stage III patients with perfusing rhythms that require aggressive external rewarming and correction of refractory acidosis without the need for cardiopulmonary support. Alternatively, VA-ECMO is appropriate for hypothermic cardiac arrest patients (Stage IV), necessitating mechanical circulatory support. ECMO utilization is a resource-intensive medical intervention that entails coordinating multiple interdisciplinary teams in prehospital and hospital settings. During data collection, it was noted that across successful vascular cannulation efforts, specific personnel, including prehospital paramedics and ED faculty, were found to be repeatedly involved in patient care.

Contraindications to ECMO initiation in normothermic cardiac arrest, including initial asystole, unwitnessed cardiac arrest, and prolonged CPR duration, are not absolute contraindications for ECMO when applied to accidental hypothermic cardiac arrest (9). Severe traumatic injuries, advance directives, and futility are among the reasons patients may not be started on ECMO in hypothermic cardiac arrests. Further considerations should include end-stage organ dysfunction and contraindications to ECMO-related anticoagulation, as demonstrated in case one, although anticoagulation free extracorporeal life support (ECLS) is feasible (9-11).

The Hypothermia Outcome Prediction after ECLS (HOPE) score can be utilized to further differentiate hypothermic ECMO patients based upon a calculated probability range from 0% to 100% chance of survival to hospital discharge (11,12). For the cases presented, a HOPE score was retrospectively calculated (Table 1). No patients in this case series met the criteria for asphyxia, which would contribute to a particularly worsened prognosis in the same patient demographic. It should be noted that the HOPE score was intended to stratify survivability using ECMO in patients in active cardiac arrest due to hypothermia; however, cases 1, 3, and 5 were post-arrest at the time of ECMO initiation.

Neurologic outcomes were retrospectively obtained utilizing the CPC scale, with a favorable outcome defined as CPC ≤2. Of note, patients who were placed on VV-ECMO had the most favorable prognosis, with a CPC score of one and disposition to home (Table 2). Vulnerable populations at increased risk for environmental cold exposure include those with unsheltered status, substance use/impairment, psychiatric illness, and the elderly (13). This trend was consistently demonstrated above with five case presentations of known/charted history of drug or alcohol use.

Study limitations

This retrospective observational case series includes a limited number of patients due to the nuanced inclusion criteria for initiation of ECMO for severe accidental hypothermia in a region not traditionally associated with cold environmental exposure. Furthermore, given the lack of a standardized protocol, selection bias concerns are present. This case series is nonconsecutive as only patients who received vascular cannulation for ECMO for severe hypothermia were included. Potential cases exist where early prehospital detection, appropriate triage, or eligible patient exclusion failed to result in vascular cannulation. The authors considered this limitation and were unable to identify cases of ECMO consideration for severe hypothermia without vascular cannulation.

Conclusions

ECMO should be considered in patients presenting to the ED with severe accidental hypothermia (T_core <28 ℃) and hemodynamic instability, as it can improve clinical outcomes. This case series demonstrates the existence of severe accidental hypothermia in the Southeastern US, a region not typically associated with such extreme cold related incidents, as well as and need for ECMO protocolization to appropriately triage and treat such presentations. These findings challenge existing paradigms and suggest that ECMO capabilities may be beneficial in areas with varying climate profiles. Although ECMO protocols for hypothermia have historically been associated with colder climates, these cases underscore that severe accidental hypothermia can and does occur in more temperate environments, and a structured ECMO program can result in good outcomes for patients. As climate variability increases and vulnerable populations remain at risk for environmental exposure, institutions in non-traditional hypothermia zones should consider the value of structured protocols and early referral pathways for suspected hypothermia with cardiac instability. Further research is required to devise protocols that appropriately triage patients in the prehospital setting to ECMO-capable medical centers. Although not utilized in the above cases, the RSHS offers promise for purely qualitative clinical assessment—an asset for early detection and triage to ECMO-capable centers—given the difficulty obtaining accurate body temperatures in the prehospital setting. Further research is also necessary to investigate the potential utilization of VV-ECMO in a subset of Stage II—hemodynamically unstable, hypothermic patients with a perfusing rhythm—as an alternative to VA-ECMO.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the AME Case Series reporting checklist. Available at https://jeccm.amegroups.com/article/view/10.21037/jeccm-2025-35/rc

Peer Review File: Available at https://jeccm.amegroups.com/article/view/10.21037/jeccm-2025-35/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jeccm.amegroups.com/article/view/10.21037/jeccm-2025-35/coif). J.P.G. receives simulation grant from SAEM and honoraria from ACCP. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All procedures performed in this study were in accordance with the ethical standards of the institutional and/or national research committee(s) and with the Declaration of Helsinki and its subsequent amendments. The publication of this case series was waived from patients’ consent, according to the Advocate Health – Wake Forest University School of Medicine Institutional Review Board.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Peiris AN, Jaroudi S, Gavin M. Hypothermia. JAMA 2018;319:1290. [Crossref] [PubMed]
2. Mazur P, Kosiński S, Podsiadło P, et al. Extracorporeal membrane oxygenation for accidental deep hypothermia-current challenges and future perspectives. Ann Cardiothorac Surg 2019;8:137-42. [Crossref] [PubMed]
3. Hymczak H, Gołąb A, Mendrala K, et al. Core Temperature Measurement-Principles of Correct Measurement, Problems, and Complications. Int J Environ Res Public Health 2021;18:10606. [Crossref] [PubMed]
4. Durrer B, Brugger H, Syme D, et al. The medical on-site treatment of hypothermia: ICAR-MEDCOM recommendation. High Alt Med Biol 2003;4:99-103. [Crossref] [PubMed]
5. Musi ME, Sheets A, Zafren K, et al. Clinical staging of accidental hypothermia: The Revised Swiss System: Recommendation of the International Commission for Mountain Emergency Medicine (ICAR MedCom). Resuscitation 2021;162:182-7. [Crossref] [PubMed]
6. Brown DA. Hypothermia. In: Tintinalli JE, Ma OJ, Yealy DM, et al., eds. Tintinalli's Emergency Medicine: A Comprehensive Study Guide. 9th ed. McGraw Hill; 2020:1337-45.
7. Prekker ME, Rischall M, Carlson M, et al. Extracorporeal membrane oxygenation versus conventional rewarming for severe hypothermia in an urban emergency department. Acad Emerg Med 2023;30:6-15. [Crossref] [PubMed]
8. Takauji S, Hayakawa M, Yamada D, et al. Outcome of extracorporeal membrane oxygenation use in severe accidental hypothermia with cardiac arrest and circulatory instability: A multicentre, prospective, observational study in Japan (ICE-CRASH study). Resuscitation 2023;182:109663. [Crossref] [PubMed]
9. Swol J, Darocha T, Paal P, et al. Extracorporeal Life Support in Accidental Hypothermia with Cardiac Arrest-A Narrative Review. ASAIO J 2022;68:153-62. [Crossref] [PubMed]
10. Lott C, Truhlář A, Alfonzo A, et al. European Resuscitation Council Guidelines 2021: Cardiac arrest in special circumstances. Resuscitation. 2021;161:152-219. Erratum in: Resuscitation 2021;167:91-92. [Crossref] [PubMed]
11. Pasquier M, Hugli O, Paal P, et al. Hypothermia outcome prediction after extracorporeal life support for hypothermic cardiac arrest patients: The HOPE score. Resuscitation 2018;126:58-64. [Crossref] [PubMed]
12. Pasquier M, Rousson V, Darocha T, et al. Hypothermia outcome prediction after extracorporeal life support for hypothermic cardiac arrest patients: An external validation of the HOPE score. Resuscitation 2019;139:321-8. [Crossref] [PubMed]
13. Mohr WJ, Jenabzadeh K, Ahrenholz DH. Cold injury. Hand Clin 2009;25:481-96. [Crossref] [PubMed]