Background and Overview

Article Title/Citation Hypothermia vs. normothermia after out-of-hospital cardiac

arrest.
Study objectives/purpose Determine if hypothermia vs. normothermia affects all-cause
mortality in out-of-hospital cardiac arrest patients.
Brief background (HYPERION trial 2018) 584 patients with non-shockable rhythm
Why issue is important, concise showed better neurologic outcomes with targeted hypothermia
summary of previous literature. at 33 C (91.4 F) vs 37 C (98.6 F) normothermia had increased
survival and improved neurologic outcome.

European Society and AHA guidelines recommend targeting

temperature management with a target between 33 and 36

TTM1 trial- 33C vs 36 C (96.8), 950 pts, results showed no

difference between the treatments.
Funding sources Grants from nonprofit and government agencies (Swedish
research council et al.)
Methods
Study design and methodology International, multicentered, parallel group superiority trial.
Study design, randomization, Randomized to 1:1 ratio using web-based system. The clinicians
blinding, controls, study groups, were not blinded, however, assessors, participants, outcome
length of study, interventions, assessors, statisticians, and data managers were blinded.
group distinctions, etc . Are these Duplicate manuscripts were used to help blinding.
appropriate?
Co-enrolment with the TAME trial (Targeted Therapeutic Mild
Hypercapnia)

1900 patients. Enrollment was from Nov. 2017-January 2020.

37 withdrawn due to lack of consent and 2 were randomized
twice.
Cooling methods included surface and intravascular
temperature management device.
Intervention
Target temperature hypothermia group of 33 C, followed by
controlled rewarming after 28 hours. Intervention period 40
hours
Control
Target temperature normothermia group > 37.8 C, with early
treatment of fever.

Patient demographics were well balanced between groups.

Slightly more HTN and bystander performed CPR in the
hypothermia group.
Patient selection and enrollment Inclusion criteria
Inclusion /exclusion criteria, sample > 18 YO, Out-of-hospital cardiac arrest of cardiac or unknown
size needed (if discussed), etc. Are cause. All were unconscious with a score of <4 on Full outline of
these appropriate based on ‘real - unresponsiveness or “FOUR score” and no verbal response to
world’? pain. > 20 min. ROSC after resuscitation
1900 patients enrolled
Exclusion criteria
Interval from ROSC to screening > 180 min. Unwitnessed
cardiac arrest w/ asystole as initial rhythm. Limitations in care.

Estimated sample size of 1862 patients with 90% power to

detect a relative deduction of 15% in the risk of death, and a
two-sided alpha level of 0.05. With an estimated ARR of 7.5
percentage points based on previous trial data
Outcome measures/endpoints Primary- Death from any cause at 6 months (all-cause
List primary and secondary. Are mortality)
these appropriate given study
objectives Secondary outcomes include functional outcome at 6 months
using the Rankin scale (0-6). A score of 4-6 means there is a
poor functional outcome. Functional outcome good/poor was
used if patients were unable to have structured assessment
(Due to COVID)
Other outcomes include days the patient was alive out of the
hospital until day 180. Survival determined in a time-to-death
analysis and health related quality of life determined by visual
analogue scale EQ-5D-5L (0-100' 100=good)

Adverse events prespecified pneumonia, sepsis, bleeding

arrhythmia resulting in hemodynamic compromise, skin
complications from device used for targeted temperature
management.
Statistical analyses Survival analysis performed with COX regression
Summarize tests done and state Use of R analysis (language and environment for statistical
whether appropriate given type o f computing)
data and study design? How did Primary analysis and dichotomous outcomes were assessed
they account for missing data? with mixed effects generalized linear model with a logit link
(adjusted for stratification variables and were reported as
population-level relative risks derived by G computation.
Results
Baseline info 1900 enrolled, 37 withdrew due to consent
Number of study subjects actually 2 patients underwent randomization twice
enrolled per group, any baseline 1861 intention to treat of 930 assigned to hypothermia and 931
differences that could have to normothermia.
affected study results, number
starting study versus completing, 1850 analyzed for survival (11 lost to follow up)
sample size attainment, etc.
No significant difference between Age, Sex, Med Hx,
Characteristics, Median time to ROSC, Temp, ST elevation MI.
Summary of study results, 1850 pts. 33 C (91.4 F) 37 C (98.6 F)
focusing on primary and 6 mo. all-cause mortality 465/925 (50%) 446/925
secondary outcomes (48%)
Including subgroup analysis, etc. Be RR 1.04 (95% CI 0.94-1.14) p = 0.37
sure to include both efficacy and Rankin Score of 4-6 at 6 mo. 488/881 (55%) 479/866
safety parameters, if available. List (94% of patients) (55%)
specific numerical and statistical RR 1.00 (95% CI 0.92-1.09)
endpoints for relevant differences AE: Arrythmias causing 222/927 (24%) 152/921
and discuss both statistical and hemodynamic compromise (16%)
clinical significance. Suggest P<0.001
making subheading based on
stated outcome
measures/endpoints
Discussion
Brief summary of authors’ main That the results were in line with the TTM trial.
discussion points
Be sure to also include any
comparisons with other
literature/studies/therapies
Study strengths and weaknesses Strengths (identified by the author):
Include strengths & limitations the Patients were cooled at a similar or faster rate than that in
authors listed, as well as your most previous trials. All sites had previous experience with the
personal evaluation of strengths & use of hypothermia. Generalizable due to large sample size,
weaknesses/limitations. This broad eligibility criteria, and number of hospitals that
section should be large and clearly participated.
demonstrate that you have
critically evaluated the article. Be Strengths (identified by the reviewer):
sure to comment on internal and Well designed. Good Internal validity with intention to treat
external validity issues with the analysis, blinding everyone except treating clinicians,
trial. 1747/1861 (94%) Rankin functional outcomes were assessed.
External validity was high due to above mentioned sample size,
number of hospitals in different countries. Approved by ethics
committees in each country as well as an independent data and
safety monitoring committee reviewed the data and preformed
and prespecified, blinded interim analysis. Paid for through
government grants.

Weaknesses/limitations (identified by the author):

Standard of care in the ICU (sedation, paralysis, and mechanical
ventilation were included in protocol. Their effect is unknown)
The trial contained conservative protocol for assessment of
neurologic prognosis and guidance for withdrawal of life
support. ICU staff were aware of the assigned target
temperature during the ICU stay. Possible confounder that 1/5th
of patients were also enrolled in the TAME trial (still ongoing)

Weaknesses/limitations (identified by the reviewer):

 Failed to meet estimated sample size of 1862 patients (actual
1850)
Conclusions
Author’s conclusions Induced hypothermia to 33 C did not lead to lower mortality in
patients that suffered an out of hospital cardiac arrest, or
functional outcome and quality of life improvement.
Evaluator/student’s conclusions I do agree with the authors conclusions. This study seems to
Do you agree with the author’s have been well designed. Hyperkalemia shift
conclusions? Why or why not?
What different/additional
conclusions do you have?
Applicability and impact on I believe the impact will be to discontinue the widespread
pharmacists/healthcare providers practice of induced hypothermia until there is more evidence
and potentially patients/public. that support its use.
Should practice change based on
this study?
Biggest take home message(s) in 3 It no longer seems work the effort to lower patients body
sentences or less. temperature to < 37 F in patients following out of hospital
cardiac arrest. Future study might look at normothermia vs no
temperature management.