Skip to content
View commentaries on this research

Please note that this summary was posted more than 5 years ago. More recent research findings may have been published.

This is a plain English summary of an original research article. The views expressed are those of the author(s) and reviewer(s) at the time of publication.

Specifically treating fever in adults with sepsis did not reduce the number of people dying within 28 days. It also had no effect on the frequency of hospital-acquired infections, reversing septic shock, lowering heart rate or improving breathing. Sepsis is a rare complication of an infection. Among the 123,000 people a year in England who develop sepsis around 30% will die from it, so improving our understanding of how to treat sepsis is clearly important.

Early treatment of sepsis is important for a number of reasons, but the findings of this review imply that it is not necessary to specifically treat fever. We don’t know whether there might be benefits for people with very high temperatures.

Why was this study needed?

Sepsis is a complication of an infection, it is rare but is serious and can lead to organ failure and even death if it is not identified and treated promptly. Around 123,000 people a year develop sepsis in England, but it is often diagnosed late and can progress very swiftly. Many require treatment in hospital, often in intensive care, and around 37,000 people die from sepsis every year.

Both high body temperature (fever) and low body temperature may be signs of sepsis. The fever associated with sepsis is sometimes treated using drugs or other treatments such as physical cooling.

The role of fever in sepsis is poorly understood and it is unclear whether fever is beneficial or harmful. Therefore there is uncertainty about whether or not to treat fever.

What did this study do?

This systematic review and meta-analysis included eight randomised controlled trials of 1,507 adults and eight observational studies of 17,432 adults.  Most studies were from the US, none from the UK.

Treatments for fever were paracetamol, nonsteroidal anti-inflammatory drugs such as ibuprofen, and physical cooling. It was not clear what doses of the drugs were given, the precise physical cooling methods used or the duration of treatment.

The randomised trials were mostly at low risk of bias, though most were small studies apart from the HEAT trial from New Zealand.  Almost all (99%) of the data for the meta-analysis of observational data on 14 day mortality came from a single large case-control study which may have had selection bias meaning that it is less reliable.

What did it find?

  • Treating fever did not reduce the number of people dying within 28 days according to the randomised controlled trials (relative risk [RR] 0.93, 95% confidence interval [CI] 0.77 to 1.13) or the observational studies (odds ratio [OR] 0.90, 95% confidence interval [CI] 0.54 to 1.51).
  • Findings for death within 14 days in the intensive care unit (ICU) were mixed, with randomised trials showing a significant reduction (RR 0.68, 95% CI 0.49 to 0.92), but observational studies suggesting a significant increase (OR 1.35, 95% CI 1.15 to 1.59). The observational data were from 2 studies, one a large case-control study with significant differences in age and severity scores, suggesting these are less reliable data.
  • Body temperature was significantly reduced following fever treatment (RR ‑0.38, 95% CI ‑0.63 to ‑0.13) based on randomised trials.
  • Treating fever did not reduce the frequency of hospital-acquired infections, orimprove the number of people who recovered from septic shock.
  • Fever treatment had no effect on heart rate or a measure of lung function and breathing rate called minute ventilation.

What does current guidance say on this issue?

NICE guidance from 2016 does not make any recommendations about whether or not to reduce fever during sepsis. They do say that sepsis should not be ruled in or out based on fever or hypothermia alone.

What are the implications?

The evidence for early death within 14 days was mixed, with randomised controlled trials indicating a reduced risk and observational studies suggesting an increased risk. The largest observational study was at risk of bias and the researchers say that this early outcome is of questionable importance and should not influence clinical practice.

A recent survey of 139 Intensive Care Units in 23 countries found that most controlled fever with paracetamol. This review suggests that contrary to current practice, it is not necessary to specifically treat fever in people with sepsis.

It is not known if people with higher temperatures might benefit from treatment of fever. The randomised trials used a cut-off of between 38.0°C to 38.4°C while the observational studies ranged between 37.3°C to 39.5°C.


Citation and Funding

Drewry AM, Ablordeppey EA, Murray ET, et al. Antipyretic Therapy in Critically Ill Septic Patients: A Systematic Review and Meta-Analysis. Crit Care Med. 2017;45(5):806-13.



NCEPOD. Just say sepsis! A review of the process of care received by patients with sepsis. London: National Confidential Enquiry into Patient Outcome and Death; 2015.

NHS Choices. Sepsis. London: Department of Health; 2016.

NICE. Sepsis: recognition, diagnosis and early management. NG51. London: National Institute for Health and Care Excellence; 2016.

Niven DJ, Laupland KB, Tabah A, et al. Diagnosis and management of temperature abnormality in ICUs: a EUROBACT investigators’ survey. Critical Care. 2013.17:R289.

Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre


NIHR Evidence is covered by the creative commons, CC-BY licence. Written content may be freely reproduced provided that suitable acknowledgement is made. Note, this license excludes comments made by third parties, audiovisual content, and linked content on other websites.

  • Share via:
  • Print article
Back to top