Publication

• Title: Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest
• Acronym: HACA
• Year: 2002
• Journal published in: The New England Journal of Medicine
• Citation: Hypothermia after Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med. 2002;346(8):549-556.

Context & Rationale

• Background

  • Comatose survivors of out-of-hospital cardiac arrest frequently died or had severe neurological disability due to global cerebral ischaemia–reperfusion injury during the post-resuscitation period.
  • Experimental and early clinical data suggested that mild hypothermia could attenuate reperfusion injury, but high-quality randomised evidence in adults was lacking.
  • Post-resuscitation fever was increasingly recognised as potentially harmful, raising the possibility that active temperature control could be a modifiable determinant of outcome.
  • Accompanying editorials framed the emerging RCT evidence as biologically plausible and practice-relevant, while emphasising the need for careful implementation and further trials. ¹
  • A contemporaneous Perspective piece highlighted hypothermia as a potentially “simple” neuroprotective strategy, but one requiring rigorous evaluation and systems-level delivery. ²
• Research Question/Hypothesis

  • In adults (18–75 years) resuscitated after witnessed out-of-hospital ventricular fibrillation cardiac arrest who remain comatose, does induced mild hypothermia (32–34°C) improve favourable neurological outcome at 6 months compared with standard normothermia?
  • Hypothesis: hypothermia would increase the proportion with Cerebral Performance Category (CPC) 1–2 and reduce mortality, without materially increasing complications.
• Why This Matters

  • Placed functional neurological recovery (rather than physiological surrogates) at the centre of post-cardiac arrest interventional research.
  • Helped establish targeted temperature management as a core component of post-resuscitation care pathways and stimulated a large subsequent trial programme.
  • Raised enduring methodological questions about fever prevention versus active hypothermia, and about generalisability beyond the highly selected ventricular fibrillation cohort.

Design & Methods

• Research Question: Among comatose adults resuscitated after witnessed out-of-hospital ventricular fibrillation cardiac arrest, does induced mild hypothermia (32–34°C) improve favourable neurological outcome (CPC 1–2) at 6 months compared with standard normothermia?
• Study Type: Multicentre, randomised, controlled trial with blinded assessment of neurological outcome; post-resuscitation ICU care following out-of-hospital cardiac arrest.
• Population:
  • Setting: Adults admitted after successful resuscitation from out-of-hospital cardiac arrest; comatose after return of spontaneous circulation (ROSC).
  • Key inclusion criteria: Age 18–75 years; witnessed arrest; initial rhythm ventricular fibrillation; interval from collapse to first basic life support attempt by bystander 5–15 min; interval from collapse to first resuscitation attempt by emergency personnel 15–60 min; ROSC within 60 min; persistent coma (no response to verbal commands) after ROSC.
  • Key exclusion criteria: Tympanic temperature on admission <30°C; coma before arrest due to CNS depressant drugs; pregnancy; response to verbal commands before randomisation; mean arterial pressure <60 mmHg for >30 min after ROSC; oxygen saturation <85% for >15 min after ROSC; terminal illness; follow-up impractical; enrolment in another study; arrest after arrival of emergency personnel; known pre-existing coagulopathy.
• Intervention:
  • Cooling method: External cooling using the TheraKool device (forced cold air over the body) with escalation to ice packs and cold (4°C) Ringer’s lactate if needed.
  • Target and monitoring: Bladder temperature 32–34°C.
  • Timing goals: Start cooling as soon as possible after ROSC and aim to reach target temperature within 4 h after ROSC.
  • Cold fluid protocol: Ringer’s lactate at 4°C, 30 mL/kg (maximum 2 L) over 30 min if target was not achieved with external cooling.
  • Sedation and shivering control: Midazolam 0.125 mg/kg/h and fentanyl 0.002 mg/kg/h (adjustable) plus pancuronium 0.1 mg/kg every 2 h for 32 h.
  • Maintenance and rewarming: Maintain 32–34°C for 24 h from the start of cooling, then passive rewarming over approximately 8 h.
• Comparison:
  • Standard care: Conventional intensive care with “normothermia”; patients placed on a conventional bed without induction of hypothermia.
  • Co-interventions: Other post-resuscitation therapies (ventilation, haemodynamic support, thrombolysis/revascularisation, etc.) delivered according to local practice and clinician judgement.
• Blinding: Treating teams were not blinded (physicians managing the first 48 h could not be blinded); physicians responsible for outcome assessment were blinded to treatment assignment.
• Statistics: Power calculation (effect size, alpha, beta/power, required sample size): Not reported. Analysis: intention-to-treat; risk ratios with 95% CIs; Kaplan–Meier survival analysis; multivariable logistic regression planned to assess confounding (age; interval from cardiac arrest to initiation of basic life support; interval from cardiac arrest to ROSC; diabetes; coronary heart disease).
• Follow-Up Period: Primary endpoints assessed at 6 months; complications tracked during the first 7 days; in-hospital outcomes also reported.

Key Results

This trial was stopped early. Enrolment ended after 275 patients were randomised due to slow enrolment and termination of funding.

• Hypothermia was associated with higher rates of favourable neurological outcome at 6 months (55% vs 39%) and lower mortality at 6 months (41% vs 55%).
• Protocol delivery achieved target temperature relatively late (median 8 h after ROSC), with 19 patients unable to reach target temperature and 14 patients requiring early discontinuation; despite this, outcomes favoured hypothermia.
• Prespecified complications were common in both groups, with similar proportions experiencing any complication (73% vs 70%; P=0.70), while pneumonia, sepsis and bleeding were numerically more frequent in the hypothermia arm.

Internal Validity

• Randomisation and allocation: Block randomisation (blocks of 10) stratified by centre using sealed envelopes; concealment integrity beyond the envelope process was not described.
• Blinding and risk of bias: Treating clinicians could not be blinded during the first 48 h; physicians assessing outcomes were blinded; unblinded delivery allows performance bias through co-interventions and decisions around limitation of life-sustaining therapy (not reported).
• Post-randomisation exclusions and missingness: Neurological outcome at 6 months was unavailable for 1 patient in each group (analysed as 75/136 vs 54/137); early discontinuation of cooling occurred in 14 patients (haemodynamic instability 8; bleeding 2; pulmonary oedema 2; admission hypothermia 1; technical problem 1).
• Baseline characteristics: Groups were broadly similar (median age 59 years in both arms; female sex 32/138 [23%] vs 33/137 [24%]); notable imbalances included diabetes 26/138 (19%) vs 11/135 (8%) and coronary heart disease 59/138 (43%) vs 43/135 (32%) (both more common in the normothermia group).
• Timing: Cooling started median 105 min after ROSC; target temperature 32–34°C reached median 8 h after ROSC.
• Dose: Target temperature 32–34°C; median duration of cooling 24 h; rewarming over approximately 8 h; 94/134 (70%) required ice packs to reach the target.
• Separation of the variable of interest: Intervention protocol explicitly targeted 32–34°C with bladder temperature monitoring; normothermia strategy was not protocolised beyond “normothermia”, and detailed quantification of temperature-time separation beyond the intervention target was not reported.
• Protocol adherence and contamination: Cooling targets were not achieved in 19 patients; crossover in the control group to active cooling was not reported.
• Outcome assessment: Primary endpoint used CPC at 6 months (CPC 1–2 favourable); outcome assessors were blinded; endpoints included objective mortality and functional neurological category.
• Statistical rigour: Intention-to-treat analysis; effect sizes reported as risk ratios with 95% CIs; multivariable logistic regression prespecified to assess confounding; no sample size/power calculation was reported, and the trial stopped early due to funding/feasibility.

Conclusion on Internal Validity: Overall, internal validity appears moderate: randomisation and blinded outcome assessment support causal inference, but open-label delivery, incomplete attainment of target temperature in a sizeable minority, absence of a reported power calculation, and early stopping increase susceptibility to performance bias and imprecision.

External Validity

• Population representativeness: Eligibility was restricted to witnessed out-of-hospital arrests with an initial shockable rhythm and relatively constrained time windows (bystander basic life support within 5–15 min; EMS resuscitation attempt within 15–60 min; ROSC within 60 min), plus haemodynamic and oxygenation stability after ROSC.
• Exclusions and spectrum: Findings do not directly address nonshockable rhythms, unwitnessed arrests, prolonged downtimes, refractory shock or sustained hypoxaemia after ROSC, pregnancy, or patients >75 years.
• Applicability and feasibility: Cooling was delivered using external forced-air cooling with escalation to ice packs and cold fluids (4°C), coupled to deep sedation and neuromuscular blockade; this is transferable to many ICUs but requires staff familiarity, temperature monitoring, and complications surveillance.
• Systems considerations: Outcomes may vary with prehospital response times, postarrest coronary reperfusion pathways, and local practices around neuroprognostication and withdrawal of life-sustaining therapy (not described).

Conclusion on External Validity: Generalisability is limited to a highly selected subgroup of comatose survivors of witnessed shockable out-of-hospital arrest with moderate downtimes and early post-ROSC stability; extrapolation to broader cardiac arrest populations requires additional evidence.

Strengths & Limitations

• Strengths:
  • Randomised controlled design with clinically meaningful functional primary endpoint (CPC at 6 months).
  • Blinded assessment of outcomes, reducing detection bias for the primary endpoint.
  • Pragmatic, ICU-feasible cooling strategy (external cooling with escalation using cold fluids and ice packs).
  • Clear, protocolised temperature target and duration (32–34°C for 24 h, rewarming over ~8 h).
• Limitations:
  • Stopped early due to funding and slow enrolment, increasing risk of imprecision and overestimation of effect.
  • Open-label delivery with potential for differential co-interventions and post-arrest prognostication/withdrawal decisions (not reported).
  • Highly selected population (witnessed shockable rhythm with constrained timing windows and early stability), limiting generalisability.
  • Delay to target temperature (median 8 h after ROSC) and incomplete achievement of targets (19 unable to reach target; 14 discontinued early) complicate interpretation of dose–response and mechanism.
  • Control group temperature management was not protocolised beyond “normothermia”, leaving uncertainty about the relative contributions of induced hypothermia versus fever avoidance.

Interpretation & Why It Matters

• Clinical signal

  In a narrowly defined shockable out-of-hospital arrest cohort, induced mild hypothermia was associated with improved functional recovery and survival at 6 months, with broadly similar overall complication rates.
• Mechanism and implementation

  The protocol achieved meaningful temperature separation despite relatively late attainment of target temperature, supporting temperature modulation as a biologically active component of post-resuscitation care.
• Modern relevance

  Later evidence has narrowed the apparent benefit of deep hypothermia relative to active normothermia with fever prevention, reframing HACA as a foundational trial demonstrating the importance of temperature control rather than establishing a single “best” target temperature for all postarrest patients.

Controversies & Subsequent Evidence

• Generalisability and selection: Correspondence highlighted that the eligible cohort represented a minority of screened postarrest patients (restrictive rhythm and timing criteria), limiting inference to broader arrest populations. ³
• Baseline neurological severity and prognostic balance: Correspondence noted limited reporting of coma depth/severity at enrolment, raising concern about residual prognostic imbalance despite randomisation. ³
• Metabolic and fluid effects of cooling: Trial correspondence provided additional protocol-level data suggesting modest potassium differences (median 3.5 vs 3.7 mmol/L at 24 h; 4.0 vs 4.4 mmol/L at 48 h, hypothermia vs normothermia) and higher administered fluid volumes with hypothermia (10.7 vs 8.8 L) with a more positive fluid balance (5.6 vs 4.2 L). ³
• Target temperature versus fever avoidance: Subsequent RCTs comparing 33°C versus 36°C did not demonstrate clear superiority of 33°C for major outcomes, challenging a “lower is better” interpretation. ⁴
• Extension beyond shockable rhythms: A later RCT in nonshockable rhythms reported improved neurological outcome with hypothermia versus normothermia, but the overall evidence base remains heterogeneous by rhythm and setting. ⁵
• Hypothermia versus active normothermia with fever prevention: A large contemporary RCT comparing hypothermia to targeted normothermia with active fever prevention did not show improved outcomes with hypothermia and reported more arrhythmias causing haemodynamic compromise in the hypothermia group. ⁶
• Guideline evolution: Modern guidance has shifted from routine induction of 32–34°C in all eligible patients to an emphasis on active temperature control and fever prevention, with selection of a target temperature (within a recommended range) tailored to context and patient factors. ⁷
• Implementation framing in North American guidance: Contemporary scientific statements emphasise temperature management as a bundle (continuous monitoring, controlled cooling/normothermia, and avoidance of fever) rather than a single “hypothermia-only” intervention. ⁸
• Evidence synthesis: Updated systematic review/meta-analysis incorporating newer trials has reduced certainty that hypothermia confers mortality or favourable neurological benefit versus active normothermia strategies. ⁹

Summary

• Multicentre RCT in comatose survivors of witnessed shockable out-of-hospital cardiac arrest showed improved favourable neurological outcome at 6 months with mild hypothermia (55% vs 39%; RR 1.40; 95% CI 1.08 to 1.81; P=0.009).
• Mortality at 6 months was lower in the hypothermia group (41% vs 55%; RR 0.74; 95% CI 0.58 to 0.95; P=0.02).
• The trial stopped early due to funding/slow enrolment, and target temperature was reached relatively late (median 8 h after ROSC) with incomplete protocol attainment in some patients.
• Overall complication rates were similar (73% vs 70%; P=0.70), with infections and bleeding numerically higher with hypothermia.
• Later evidence and guidelines have reframed the key message as the importance of active temperature control and fever prevention, with uncertainty about routine deep hypothermia for all patients.

Further Reading

Other Trials

• 2002Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med. 2002;346(8):557-563.
• 2013Nielsen N, Wetterslev J, Cronberg T, et al. Targeted temperature management at 33°C versus 36°C after cardiac arrest. N Engl J Med. 2013;369(23):2197-2206.
• 2019Lascarrou JB, Merdji H, Le Gouge A, et al. Targeted temperature management for cardiac arrest with nonshockable rhythm. N Engl J Med. 2019;381(24):2327-2337.
• 2021Dankiewicz J, Cronberg T, Lilja G, et al. Hypothermia versus normothermia after out-of-hospital cardiac arrest. N Engl J Med. 2021;384(24):2283-2294.

Systematic Review & Meta Analysis

• 2023Granfeldt A, Holmberg MJ, Andersen LW, et al. Temperature control after adult cardiac arrest: an updated systematic review and meta-analysis. Resuscitation. 2023;191:109928.
• 2022Aneman A, Cronberg T, Nolan JP, et al. Temperature management in adult cardiac arrest: systematic review and (Bayesian) meta-analysis. Critical Care. 2022.
• 2021Fernando SM, Di Santo P, Sadeghirad B, et al. Targeted temperature management following out-of-hospital cardiac arrest: a systematic review and network meta-analysis. Intensive Care Med. 2021.
• 2023Duhan M, Raza S, Waqas M, et al. Hypothermia compared to normothermia following out-of-hospital cardiac arrest: a systematic review and meta-analysis. Am J Cardiol. 2023.
• 2023Arrich J, Zeiner A, Sterz F, et al. Hypothermia for neuroprotection in adults after cardiac arrest. Cochrane Database Syst Rev. 2023.

Observational Studies

• 2023Haugk M, Deakin CD, Lall R, et al. Comparative before–after study of fever prevention versus targeted temperature management after out-of-hospital cardiac arrest. Resuscitation Plus. 2023;100538.
• 2021Garfield B, MacFie J, Shah R, et al. Temporal changes in targeted temperature management for survivors of cardiac arrest: compliance with recommended treatment and 90-day neurologic outcomes. Ther Hypothermia Temp Manag. 2021.
• 2024Holgersson J, Ala-Kokko T, Tirkkonen J, et al. Targeted temperature management and fever after out-of-hospital cardiac arrest: post hoc analysis of the FINNRESUSCI study. Acta Anaesthesiol Scand. 2024.
• 2024Caddell AJ, Figueroa BE, Van Laar D, et al. Targeted temperature management for patients after cardiac arrest in contemporary practice: observational experience and outcomes. J Intensive Care Med. 2024.

Guidelines

• 2022Nolan JP, Sandroni C, Böttiger BW, et al. European Resuscitation Council and European Society of Intensive Care Medicine guidelines on temperature control after cardiac arrest in adults. Resuscitation. 2022.
• 2022Wyckoff MH, Singletary EM, Soar J, et al. 2022 International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science With Treatment Recommendations. Circulation. 2022.
• 2023Perman SM, Stanton E, Soar J, et al. Temperature Management for Comatose Adult Survivors of Cardiac Arrest: A Scientific Statement From the American Heart Association. Circulation. 2023.
• 2025Nolan JP, Sandroni C, Böttiger BW, et al. European Resuscitation Council and European Society of Intensive Care Medicine guidelines for post-resuscitation care. Resuscitation. 2025;198:110809.
• 2025American Heart Association. 2025 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care: Post–Cardiac Arrest Care. Circulation. 2025.

Notes

• Temperature management practice has evolved from routine induction of 32–34°C in all eligible patients to prioritising continuous temperature monitoring and active prevention of fever, with target selection individualised according to patient context and local protocol.

Overall Takeaway

HACA is a landmark post-resuscitation RCT because it linked an ICU-deliverable physiological intervention—mild hypothermia—to improved patient-centred outcomes (functional recovery and survival) at 6 months in a defined shockable out-of-hospital arrest cohort. While later trials and contemporary guidelines have narrowed enthusiasm for universal deep hypothermia, HACA’s enduring legacy is the demonstration that active temperature control (and avoidance of hyperthermia) is a modifiable determinant of neurological outcome after cardiac arrest.

Bibliography

• 1Safar P, Kochanek PM. Therapeutic hypothermia after cardiac arrest. N Engl J Med. 2002;346(8):612-613.
• 2Curfman GD. Hypothermia to protect the brain. N Engl J Med. 2002;346(8):606-608.
• 3Therapeutic hypothermia after cardiac arrest. N Engl J Med. 2002;347(1):63-65.
• 4Nielsen N, Wetterslev J, Cronberg T, et al. Targeted temperature management at 33°C versus 36°C after cardiac arrest. N Engl J Med. 2013;369(23):2197-2206.
• 5Lascarrou JB, Merdji H, Le Gouge A, et al. Targeted temperature management for cardiac arrest with nonshockable rhythm. N Engl J Med. 2019;381(24):2327-2337.
• 6Dankiewicz J, Cronberg T, Lilja G, et al. Hypothermia versus normothermia after out-of-hospital cardiac arrest. N Engl J Med. 2021;384(24):2283-2294.
• 7Nolan JP, Sandroni C, Böttiger BW, et al. European Resuscitation Council and European Society of Intensive Care Medicine guidelines on temperature control after cardiac arrest in adults. Resuscitation. 2022.
• 8Perman SM, Stanton E, Soar J, et al. Temperature Management for Comatose Adult Survivors of Cardiac Arrest: A Scientific Statement From the American Heart Association. Circulation. 2023.
• 9Granfeldt A, Holmberg MJ, Andersen LW, et al. Temperature control after adult cardiac arrest: an updated systematic review and meta-analysis. Resuscitation. 2023;191:109928.