(Last Updated On: January 11, 2016)


Nurses, families, and consultants often request that we treat our patients’ fevers. We often reflexively answer, “Yeah, sure, give ’em a gram of acetaminophen, why not.” Occasionally we might take the high road, instead replying “Why would I blunt the body’s response to infection, let it fight!”

Is there a benefit to calming the inflammatory response or a benefit to letting the body maintain its natural role in immunity?

Young, Paul et al. Acetaminophen for fever in critically ill patients with suspected infection. NEJM 373.23 (2015): 2215-2224.


  • Prospective, Randomized, Blinded, Controlled Trial
  • 23 ICUs in New Zealand/Australia, 700 ICU level patients with fever (>38C) and known infection
  • 2 arms, either 1gm IV acetaminophen or placebo dextrose solution q 6 hours
  • Medication stopped with: ICU discharge, Fever Resolution, Antibiotic discontinuation, or death



  • Number of ICU free days up to day 28 of study


  • Mortality at days 28 and 90
  • Survival time (number days alive) from start to day 90
  • ICU and hospital LOS
  • Hospital Free days
  • Days free from ventilator
  • Days free from inotropes/vasopressors
  • Days free from renal replacement therapy
  • Days in ICU that were free from support (free from vent/inotropes/pressors/RRT for whole day)


Screen Shot 2015-12-20 at 3.59.18 PM Screen Shot 2015-12-20 at 3.58.43 PM

Subgroup differences
It is worth noting that though mortality was unchanged between groups, the acetaminophen group had prolonged ICU and overall hospital length of stays before death. Did acetaminophen administration prolong mortality, and and is this clinically relevant? This provokes discussion as to the significance of a prolonged hospital or ICU length of stay if mortality is unchanged.

Screen Shot 2015-12-20 at 3.58.36 PM

Limitations of study

  • Did not not control for use of acetaminophen before or after study enrollment
  • Acetaminophen was administered in the IV formulation, a route we rarely use
  • Limited duration of acetaminophen usage


  • Primary outcome (ICU free days): No significant difference, absolute difference of 1 day
  • Secondary outcomes: No significant difference

Application to our practice

In a multi-center, randomized, blinded, controlled trial, we did not see any benefit or harm from giving acetaminophen. Given the lack of difference in primary and secondary outcomes, the administration of acetaminophen to the critically ill, septic patient remains a case by case and provider by provider decision.

This study did not address hyperthermia control with cooling measures. In this study, patients given acetaminophen that died had longer ICU and hospital courses, which has also been shown in some trials with patients that were cooled to normothermia. It appears that these practices may be prolonging lives, increasing costs without any benefit.

It is reasonable to assume the study population from New Zealand and Australia is applicable to our patients in the US, but we should restrict our application of these findings to critically ill, septic patients given the inclusion criteria of this study.

My practice

I will continue to avoid giving antipyretics to my febrile, sick, septic patients in the ER and ICU, but if a nurse, consultant, or attending is adamant about antipyretic administration, I will not argue with them.