Journal club 8/9: HEAT trial

The article

Young PJ, Saxena MK, Bellomo R, Freebairn RC, Hammond NE, van Haren FM, et al. The HEAT trial: a protocol for a multicentre randomised placebo-controlled trial of IV paracetamol in ICU patients with fever and infectionCrit Care Resusc. 2012 Dec;14(4):290-6

What’s the question we’re looking at? Why do we care in CCM?

Should we routinely treat fever in ICU patients?


There is some light evidence that reducing fever in certain patient populations may improve surrogate outcomes. On the other hand, there is bench data and biologic plausibility suggesting that fever is a physiologic response to infection and should be tolerated. What to do?

What type of study was this?

Multicenter, blinded, randomized controlled trial

Who was the study population? What were the inclusion/exclusion criteria?

Patients from 2013-2014 in 23 med/surg ICUs in Australia/New Zealand


  • 16 years or older
  • Temperature 38c or greater within the 12 hours before enrollment


  • Receiving antimicrobials for known or suspected infection


  • Brain injury, liver disease, pharmacologic fever (this is a trial of INFECTIOUS fevers)
  • Need for anti-inflammatories or therapeutic hypothermia
  • Moribund (death expected in 24hr)
  • Significant rhabdomyolysis (concern that fever could cause muscle damage)
  • Pregnant
  • Previously eligible and not enrolled, or transferred from another ICU where they eligible but not enrolled

What was the intervention? What was the control?

Patients block-randomized by an encrypted Web system to receive either:


1g IV acetaminophen scheduled every 6 hours


An infusion of D5W every 6 hours


Both packaged in indistinguishable 100ml bottles

Therapy continued until:

  • Fever resolves
  • 28 days pass
  • Discharged from ICU or died
  • Antimicrobials stopped
  • or developed contraindication to acetaminophen

“Rescue” physical cooling permitted if temp rose above 39.5c.

Were patients blinded? Providers? Evaluators/analysts?

Yes, all three.

What is the PRIMARY outcome?

ICU-free days, up until 28 days.

Since death ends your ICU stay, this is a composite endpoint including both ICU stay and mortality.

Planned subgroup analyses:

  • Presence/absence of septic shock
  • Use/nonuse of aspirin
  • Presence/absence of high fever (>39)
  • Where fever developed (community, inpatient, ICU)

What were the SECONDARY outcomes?

  • 28-day mortality
  • 90-day mortality
  • Days alive at 90 days
  • ICU and hospital length of stay
  • Hospital-free days
  • Vent-free days
  • Inotrope/vasopressor free days
  • RRT-free days
  • Days in the ICU free from “support” (any of the above)

Physiologic/lab outcomes:

  • Mean/max temps (axillary)
  • Proportion of patients who stopped the drug due to liver dysfunction
  • Mean CRP
  • Proportion of patients with CK >5000
  • Highest creatinine in first 7 days

Were the groups initially similar?

Roughly similar.

Other than the therapy under investigation, is there any reason why patients might have been treated differently in the two groups, and was this controlled? If not, what effect might it have?

No, except that after the study period ended (i.e. once fever ended, left ICU, etc), open-label use of acetaminophen was allowed even before the follow-up period ended.

Were there any losses/failures after enrollment, and if so, were they analyzed using intention-to-treat?

A few people received the wrong therapy or were mistakenly ineligible. Full intention-to-treat except exclusion of 10 patietns who withdrew consent.

What are the results? Are they statistically significant? Clinically meaningful? Are they the same primary/secondary outcomes initially described? Are they plausible?

700 patients enrolled

Most received about 8 doses

Acetaminophen group had about a .2c lower peak temp and .3c lower average temp

There was sustained resolution of fever in 22.8% of acetaminophen group versus 16.9% of placebo (significant).

About the same number of patients in each group had to stop drug due to liver dysfunction.

Primary outcome

No difference in ICU-free days

Secondary outcomes

  • No mortality differences at 28- or 90-days
  • No differences in ICU or hospital length of stay.
  • Acetaminophen may have been associated with shorter ICU stay among survivors (about 1 day less) and longer ICU stay among non-survivors (6 days more), which was significant

Who sponsored the study? Any reason to suspect bias? Conflicts of interest among the authors?

Funded by Health Research Council of New Zealand, the Australian and New Zealand Intensive Care Foundation, and the Waikato Medical Research Foundation, basically all public money.

Myburgh had money from Baxter, Fresenius, and CSL Bioplasma, but no apparent links to Mallinckrodt (maker of Ofirmev), not that this result would have benefitted them anyway.

Our take-home

  • Acetaminophen lowers fevers, although not very much (usually <.5c).
  • Does Tylenol for fever help outcomes? Probably not in general.
  • Could there be other subgroups for whom it does matter (e.g. neuro patients)? Don’t know.
  • Does it shorten ICU stay in survivors and prolong it in those doomed to die? Don’t know, interesting finding.
  • What to do: treat fever for comfort if needed. Otherwise, truly, who cares. Like mild hypertension, from an outcome perspective, this mostly seems to be a nuisance issue.