2025 SCCM指南：热射病的治疗

指南麻醉与急重症· 急诊与复苏 · 2025

RESEARCH AGENDA

Future research is needed on several topic areas covered by these guidelines. A summary of research priorities is presented in Table 2 .

ACKNOWLEDGMENTS

TABLE 2.

Research Agenda

Research question

Cooling strategy What is the most optimal rate of cooling? What is the most desirable time to reach target temperature?

What is the best combination of cooling methods if cold-water immersion is not feasible?

What is the role of novel cooling devices such as intranasal cooling devices, endovascular cooling devices, blankets engineered from heat-conductive biomaterials, etc?

Medications

What is the role of antimicrobial prophylaxis?

What is the role of seizure prophylaxis?

Can other medications be used to prevent organ failure (e.g., N-acetylcysteine)?

What is the role of immunomodulators to treat the inflammatory response?

Other What is the role of biomarkers to identify organ failure and prognosticate outcomes?

Can artificial intelligence be used to systems that create rapid, predictable cooling and minimize adverse effects?

Can precision medicine be used to identify at higher risk for heat stroke?

References

EN_PAPER

Society of Critical Care Medicine Guidelines for the Treatment of Heat Stroke

Authors

Keywords

cold-water immersion, cooling methods, emergency medicine, heat stroke, heat-related illness, intensive care unit Patient, Intervention, Comparison, and Outcomes Questions Patient, Intervention, Comparison, and Outcomes

RATIONALE:

OBJECTIVES:

DESIGN:

METHODS:

RESULTS:

CONCLUSIONS:

Abstract

Predicted increases in heat-related weather phenomena will result in increasing heat exposures and heat injuries, like heat stroke. Prompt recognition, early intervention, and evidence-based management are necessary to optimize outcomes.

The objective of these guidelines was to develop evidence-based recommendations for the treatment of patients with heat stroke.

The Society of Critical Care Medicine convened a multidisciplinary panel of 18 international clinicians, comprising expertise in critical care, emergency medicine, neurocritical care, surgery, trauma/burn surgery, sports medicine, athletic training, military medicine, nursing, pharmacy, respiratory therapy, and one patient representative. The panel also included a guidelines methodologist specialized in developing evidence-based recommendations in alignment with the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) methodology. Conflict-of-interest policies were strictly followed during all phases of guidelines development including panel selection and voting.

The panel members identified Patient, Intervention, Comparison, and Outcomes questions in two main areas: cooling modalities and medications that affect temperature. A systematic review for each question was conducted to identify the best available evidence, statistically analyze the evidence, and assess the certainty of the evidence using the GRADE methodology. The GRADE evidence-to-decision framework was used to formulate the recommendations. Good practice statements were included to provide additional clinical guidance.

The panel generated two strong recommendations, five good practice statements and one "only-in-the-context of research" statement. Active cooling measures are recommended over passive cooling methods, with cold-or ice-water immersion achieving the fastest cooling rate. This method should be prioritized where available. In heat stroke patients, there is no evidence to support pharmacological interventions that affect temperature control and they should be avoided.

The guidelines task force provided recommendations for the management of patients with heat stroke. These recommendations should be considered along with the patient's clinical status and available resources.

H

eat-related illness is a leading cause of weather-related fatalities and its prevalence is expected to increase (1) . Heat stroke, the most severe form of heat-related illness, is defined as a core temperature greater than 40°C presenting with CNS abnormalities. Heat stroke can be accompanied by burn injury, caused by direct contact with pavement or other surfaces, which can reach temperatures in excess of 71°C (2) . Heat stroke is subclassified as either classic (i.e., nonexertional) or exertional and can affect anyone. Certain populations though are at higher risk including infants, elderly, e491 athletes, domestically displaced, and outdoor workers such as firefighters, construction workers, landscapers, and military personnel (3) .

The increasing prevalence of heat stroke has emphasized the need for timely critical care and emergency services (4). Rapid cooling is an essential intervention to restore normal physiologic activity and minimize morbidity and mortality (5) . Cold-water immersion is frequently cited as the preferred modality, but coldwater immersion may not be feasible in some settings or when mass casualty events occur, raising questions on the best alternative strategy (6) . Once cooling has been achieved, the management of heat stroke is often centered around maintenance of organ support with minimal direct evidence to guide clinical decision-making.

Building on prior work conducted by the Society of Critical Care Medicine (SCCM) (6) , a multidisciplinary panel was commissioned to develop evidence-based guidelines for the treatment of heat stroke highlighting recommendations for critical care management that occurs across a continuum of settings.

METHODOLOGY

Intended Audience

The intended audience for these guidelines is all individuals who provide care for patients with heat stroke in both the pre-hospital and acute care settings. This includes physicians, advanced practice providers, pharmacists, nurses, respiratory therapists, medical and nursing directors, athletic trainers, coaches, and other allied health and/or administrative professionals.

Composition of the Guidelines Development Group

The SCCM commissioned an international panel of experts in critical care, emergency medicine, neurocritical care, surgery, trauma/burn surgery, sports medicine, athletic training, military medicine, nursing, pharmacy and respiratory therapy, and plus one patient representative. Practice locations included pre-hospital, emergency, and inpatient settings. A methodologist from the Guidelines in Intensive Care Development and Evaluation group of McMaster University assisted with the development of the guidelines.

Conflict of Interest Policy

Members of the panel were required to disclose conflicts of interest (COI) per the SCCMs COI policy. COI was assessed at each phase of the guidelines process and at every panel meeting. Panel members with a COI were excluded from voting on a specific recommendation if a COI was present. There were no COIs that prohibited participation in discussion or voting. There was no input from industry to produce these guidelines.

Formulating Recommendations

The panel prioritized Patient, Intervention, Comparison, and Outcomes (PICO) questions based on those that had important implications for clinical practice and evidence was available. The specific population of interest were patients with heat stroke as opposed to other less severe forms of heat-related illness (e.g., heat exhaustion, heat syncope, etc). The PICO selection process is presented in Table S1 (http://links.lww.com/CCM/ H657). The panel also rated the relative importance of each outcome to determine which outcomes were critical for decision-making. A summary of searched outcomes for all PICO questions is presented in Table S2 (http://links.lww.com/CCM/H657). The final PICO questions selected (n = 7) are listed in Table 1 .

After PICO questions were finalized, an extensive review of the scientific literature was performed from inception (of each database) through May 2024 to retrieve articles that addressed the PICO questions. The search was conducted by a medical librarian using MEDLINE, Embase, Cochrane Central Register of Controlled Trials (CENTRAL), ClinicalTrials.gov, and World Health Organization International Clinical Trials Registry Platform (search queries are provided in the Supplemental Materials, http://links.lww.com/CCM/H657). The search strategy included all ages and did not restrict to subpopulations. Search results were then uploaded to Covidence (Covidence systematic review software, Veritas Health Innovation, Melbourne, VIC, Australia) (7) for screening, full-text review, and data extraction, following the procedure for systematic review.

Clinical data were systematically reviewed for each PICO question. This was done by performing de novo systematic reviews or updating recently published high-quality reviews from peer-reviewed journals if they were available. Meta-analyses were performed using a random effects model with RevMan, Version (8) . Data that could not be quantitatively pooled or outcomes with insufficient studies to analyze were addressed narratively.

The clinical practice recommendations were then developed according to the Grading of Recommendations, Assessment, Development, and Evaluation process (9) . The Evidence to Decision framework was completed by the panel using GRADEpro software (GRADEpro GDT: GRADEpro Guideline Development Tool [Software] . McMaster University and Evidence Prime, 2024) (10) for each PICO to develop a draft recommendation considering the balance of desirable and undesirable effects, certainty of the evidence, resource considerations, feasibility, acceptability, and equity considerations (Supplemental Materials, http://links. lww.com/CCM/H657, for evidence to decision worksheets). Recommendations had to receive at least 80% of the vote of the panel to be approved.

The recommendations in these guidelines are based on the balance of desirable and undesirable effects that should meet the needs of most patients in most situations. Remarks are provided, as necessary, to guide clinicians in the implementation of these recommendations. Each recommendation statement was assigned a strength ("strong" or "weak"). A "strong" recommendation is one that clinicians should follow for almost all patients (i.e., something that might qualify as a quality measure). A "weak" or "conditional" recommendation reflects a lower degree of certainty in the appropriateness of the patient care strategy for all patients. It implies that not all individuals will benefit from the recommended intervention and necessitates a careful assessment of the individual circumstances, values, and preferences. Several factors can determine the strength and direction of a recommendation including the balance between desirable and undesirable effects, confidence in the magnitude of estimates of effect on important outcomes, and confidence in values and preferences and their variability and resource use. In some situations, it may be appropriate to have a strong recommendation despite low or very low confidence in effect estimates (e.g., a circumstance where low quality evidence suggests benefit in a life-threatening situation) (9) . Guidelines panels may also make good practice statements and only-in-thecontext of research statements. There were insufficient data directly comparing active vs. passive cooling methods on clinical outcomes such as survival, length of stay, organ dysfunction, and complications. Based on limited evidence from a systematic review (3), mortality ranged from 0% to 26.5% in patients with exertional heat stroke who were treated with various cooling methods. For patients with classic heat stroke, mortality ranged from 0% to 71.4% with intensive cooling treatment (3). The review identified two pediatric case series that reported mortality rates of 25% (n = 7) and 60% (n = 50) despite treatment (3) .

The results of the subgroup meta-analyses confirmed ice-water immersion (1-5°C) and cold-water immersion (9-12°C) as the most effective cooling approach with a difference in mean cooling rate of 0.14°C/min (95% CI, 0.08-0.21°C/min) and 0.11°C/min (95% CI, Figure 1 . Mean weighted cooling rates by cooling method. To convert Celsius to Fahrenheit multiply by 1.8 and add 32. e495 0.07-0.14°C/min), respectively, compared with passive cooling (Fig. S1 , http://links.lww.com/CCM/H657). A ranking of the relative effectiveness of various active cooling methods is presented in Figure 1 . These rates may be influenced by the use of experimental designs that do not use the same initial temperatures that are often observed in clinical practice (for ethical and safety reasons) and thereby result in different temperature differentials.

There are numerous case reports describing the benefit of cold-water immersion and this method has been widely endorsed (46) (47) (48) (49) (50) . Cold-water immersion, although may be associated with adverse effects (e.g., ice directly applied to skin can lead to frostbite), introduce hazards for hospital staff (e.g., falls), interfere with ongoing resuscitation and procedures and complicate patient monitoring (e.g., cardiac monitor leads falling off). Techniques for performing cold-water immersion have been described (50) . Nevertheless, the panel determined that the desirable effects of this modality far outweighed the undesirable effects. Cold-water immersion may not be an option in some settings. Secondary alternatives are highlighted in Figure 1 . However, cold-water immersion can be successfully implemented in circumstances where mass casualty events might be expected (e.g., sporting events, road-races, etc). Careful planning and foresight are needed so required resources are readily available.

Good Practice Statement 2. Clinicians may use similar cooling strategies for either classic or exertional heat stroke.

Experimental models comparing cooling methods are primarily conducted in climate-controlled exercise laboratories using healthy volunteers. Initial temperatures do not reach the extreme values often Green reflects a final core temperature less than 39°C (preferred), yellow reflects a final core temperature between 39°C and 40°C (borderline), and red reflects a final core temperature greater than 40°C (inadequate). To convert Celsius to Fahrenheit multiply by 1.8 and add 32.

encountered in practice and patient comorbidities are often absent. Comparative trials in patients with classic heat stroke are lacking due to the ethical challenges with this experimental design. Given the harmful physiologic sequelae associated with extremes in body temperature, regardless of the root cause (excessive heat gain vs. inability to dissipate heat), the panel decided similar treatment approaches should be sought.

Good Practice Statement 3. Clinicians should choose cooling methods that reach the target temperature within 30 minutes from recognition of heat stroke symptoms.

Direct clinical evidence assessing outcomes in heat stroke patients based on the time to reach target temperature is lacking. This is largely due to the ethical challenges with performing a randomized controlled trial in this setting. One retrospective study, which reported cooling rates using a variety of modalities in 143 patients, reported mortality rates of 6.9% when a temperature of 39°C was reached in 30 minutes vs.10.6% when it was not (p = not significant) (51) . A second retrospective study of 39 patients noted a 55% relative reduction in mortality when a temperature less than or equal to 38.9°C was reached within 1 hour (p = 0.18) (52) . Two large case series reported zero fatalities when target temperatures were reached within 30 minutes (45, 53) . Despite the limited direct evidence, the degree and duration of hyperthermia are known to adversely affect normal physiologic processes. The panel decided a time to target temperature within 30 minutes is an appropriate goal.

Good Practice Statement 4. Clinicians should prioritize cooling modalities that achieve a cooling rate greater than or equal to 0.155°C/min.

Data directly comparing cooling rates on clinical outcomes are limited. One review combined case series data and found higher mortality (16% vs. 0%) with insufficient cooling ( < 0.15°C/min) and a higher rate of medical complications (risk ratio [RR], 4.57; 95% CI, 3.42-6.28) (46) . A lack of comparative trials is attributed to prohibitive ethical concerns with conducting a randomized controlled trial in which aggressive cooling is delayed. Nevertheless, the panel decided the desirable effects of aggressive cooling outweigh the undesirable effects. Additionally, when considering the goal of achieving a target temperature within 30 minutes, cooling methods with a temperature reduction rate greater than or equal to 0.155°C/min will accomplish that goal across a wide range of initial core temperatures. We recognize that cooling strategies that achieve this rate of temperature reduction may not be feasible or available in some settings. Clinicians are advised to select the method or combination of methods that will achieve the fastest rate of temperature reduction.

Medications That Affect Temperature Control

Recommendation 2. We recommend against the use of dantrolene in patients with heat stroke (strong recommendation, very low certainty of evidence).

Dantrolene is a treatment option for patients with malignant hyperthermia and its use in heat stroke has garnered some attention due to overlap in pathophysiologic mechanisms. Three randomized controlled trials have compared dantrolene with placebo in heat stroke patients (54) (55) (56) . Of those, one is only available in abstract form. Meta-analyses have demonstrated no reduction in the risk of mortality (three trials; RR, 1.01; 95% CI, 0.11-9.31; and Fig. S2 , http://links.lww. com/CCM/H657), a mean difference in cooling time (two trials) of -10.5 minutes (95% CI, -28.5 to 7.6 min; and Fig. S3 , http://links.lww.com/CCM/H657), a difference in hospital length of stay (one trial) of 1.85 days (95% CI, 0.99-2.70 d; and Fig. S4 , http://links. lww.com/CCM/H657), no difference in the risk of adverse events (two trials; RR, 0.99; 95% CI, 0.46-2.15; and Fig. S5 , http://links.lww.com/CCM/H657), and a difference in incidence of recovery of consciousness as defined by a Glasgow Coma Score of greater than or equal to 13 points at less than or equal to 90 minutes (one trial; RR, 2.5; 95% CI, 0.56-11.16; and Fig. S6 , http://links.lww.com/CCM/H657). The overall certainty of evidence was very low due to risk of bias, imprecision, and inconsistency (Table S5 , http://links. lww.com/CCM/H657).

引用 / 出处

2025 SCCM指南：热射病的治疗. 2025.