Paracetamol in Critical Illness

OVERVIEW

Paracetamol is an ubiquitous analgesic and antipyretic agent

• It is commonly used in ICU patients:
  • with fever and known or suspected infection
  • for analgesia
  • for temperature management in non-infectious conditions (e.g. stroke and TBI)
• Until the HEAT trial (2013) there were no randomised controlled trials of paracetamol in the critically ill to determine its safety and efficacy in this population

RATIONALE

Paracetamol is a widely used analgesic

• most commonly used analgesic in the world, with high clinician familiarity
• may help decrease suffering of ICU patients and limit the amounts of additional analgesia and sedation required by critically ill patients, and their associated adverse effects

Paracetamol may be beneficial if it reduces potentially harmful physiological stress by limiting fever

• potentially deleterious effects of fever
  • increased metabolic rate, heart rate, cardiac work and catecholamine production
  • At T ≥ 41C: neutrophils and macrophages show reduced immune function
• Fever is associated with worse outcomes following neurological injury
  • e.g. stroke, post-cardiac arrest hypoxic-ischaemic encephalopathy, traumatic brain injury

However, fever in critical illness may be a protective, evolutionary conserved mechanism

• Fever inhibits the function of infectious organisms
  • e.g Influenza virus, Streptococcus pneumoniae and Neisseria meningitidis
• At T38C – 40C: neutrophil, lymphocyte and macrophage immune function is enhanced
• Fever is associated with a decrease in mortality in an observational study of 600,000 ICU patients (Young et al, 2012)
• Paracetamol can prolong infections with varicella, malaria and rhinovirus (shown in non-ICU patients)(Jeffries et al, 2015)

Paracetamol may have beneficial effects unrelated to temperature control or analgesia, e.g. immunomodulation

PROS

• May help reduce physiological stress by limiting fever
• Familiar, widely used agent
• Well described safety profile (at least in non-critically patients)
• Can be given by multiple routes (PO, NG, NJ, rectal, IV) – inexpensive (except for the IV formulation)
• Provides analgesia
  • when used as part of multi-modal analgesia can decrease opioid requirements by up to 20% in non-ICU patients (Jeffries et al, 2012)
• May shorten ICU length of stay in survivors of infection (HEAT trial)

CONS

• The febrile response to infections may be adaptive
  • despite significant metabolic cost it is conserved across a broad range of species suggesting it has evolutionary advantage
  • the febrile response has been used as therapeutic intervention with success in a range of conditions
    • e.g. malaria innoculation as therapy for neurosyphilis
    • e.g. fever therapy for gonorrhea
• Analgesia
  • little evidence of analgesic efficacy in the critically ill
  • not shown to reduce opioid side-effects or improve patient satisfaction when used as part of multi-modal analgesia in non-ICU patients (Jeffries et al, 2012)
• IV paracetamol
  • more expensive than oral tablets (~AUS$6 versus ~4 cents for 1g)
  • associated with hypotension on administration
• Anti-pyresis
  • Has limited temperature reduction effect (~0.25C in the HEAT trial)
  • Ineffective at reducing temperature in TBI (PARITY trial)
• Potential for drug interactions and adverse effects
• May prolong ICU length of stay of non-survivors of infection (HEAT trial)

EVIDENCE

HEAT trial – Young et al, 2015

• Study design
  • multi-centre, blinded, randomised, controlled trial
  • n=690 patients from 23 adult ICUs in Australian and New Zealand ICUs
  • an additional 10  patients withdrew consent
• Inclusion criteria
  • ICU patients aged ≥ 16 years, temperature of ≥ 38°C within 12 hours prior to enrolment and had known or suspected infection treated with antimicrobials
• Intervention and comparison
  • 1 g of IV paracetamol q6h (n=346), versus
  • placebo (5% dextrose in water in identical glass bottles IV q6h (n=344)
    • The intervention/ control was continued until day 28, discharge from ICU, resolution of fever (temperature < 37.5°C for 24 hours), cessation of antimicrobials or death
    • Physical cooling was optional as a rescue therapy if T ≥ 39.5°C
    • NSAID use (other than aspirin) was contraindicated
    • Open label paracetamol use was permitted after the course of study medication was completed
• Outcomes
  • Primary outcome: no difference
    • number of ICU-free days (alive and discharged from ICU) at 28 days
    • 23 and 22 days in the paracetamol and placebo groups respectively (absolute difference, 0 days; 96.2% CI, 0 to 1; P = 0.07)
  • Secondary outcomes
    • no difference in ICU free-days of pre-specified subgroups, 28- or 90- day mortality, or survival time to day 90
    • heterogeneity of response with paracetamol associated with:
      • shorter ICU length of stay in survivors (3.5 days [IQR, 1.9 to 6.9] vs. 4.3 days [IQR, 2.1 to 8.9], P = 0.01)
      •  longer ICU length of stay in non-survivors (10.4 days [IQR, 4.1 to 16.9] vs. 4.0 days [IQR, 1.7 to 9.4], P < 0.001)
    • lower temperature with paracetamol
      • lower mean daily peak body temperature (38.4 ± 1.0°C vs. 38.6 ± 0.8°C; absolute difference, −0.25°C; 95% CI, −0.38 to −0.11; P < 0.001)
      • lower mean daily average body temperature (37.0 ± 0.6°C vs. 37.3 ± 0.6°C; absolute difference, −0.28°C; 95% CI, −0.37 to −0.19; P < 0.001)
• Commentary
  • Lungs were most common site of presumed infection
  • Patients who died were assumed to have zero ICU-free days
  • Strengths
    • Good baseline balance with similar peak temperatures in the 12h before randomisation: T38.8C ± 0.6°C and T38.7C ± 0.6°C in the paracetamol and placebo groups, respectively
    • Excellent follow up rates (1.4% loss to follow up)
    • Pre-published statistical analysis plan
    • Blinded and had allocation concealment
    • All analyses, including post-hoc anayses, were done prior to unmasking of data
    • Statistical analysis plan was reported before the interim analysis was conducted
    • 80% power to detect a difference of 2.2 ICU-free days in the 28 days after randomisation, at an alpha level of 0.05
  • Criticisms
    • The total paracetamol dose that patients received was low at a median of 8g (IQR 5 – 14) and the treatment duration was short (stopped at ICU discharge in ~45% and for resolution of temperture in 22.8% of the paracetamol group and 16.9% of the placebo group (odds ratio, 1.45; 95% CI, 0.99 to 2.12; P = 0.05))
    • Open label paracetamol use in both groups was high at ~ 30%, which may have diluted the treatment effect
    • Unclear how many patients had paracetamol prior to randomisation
• Conclusion
  • Paracetamol has a modest anti-pyrexial effect in ICU patients, and appears safe. Use of paracetamol may shorten ICU length of stay in survivors with infection.

PARITY trial – Saxena et al, 2015

• small RCT
• n = 41 pateints with acute traumatic brain injury
• intervention / comparison
  • 6g of IV paracetamol per day (n = 21) versus  placebo (n = 20) for 72 hours
• no difference in the mean temperature between the two groups (37.4 ± 0.5 °C in the paracetamol group vs. 37.7 ± 0.4°C in the placebo group (absolute difference -0.3°C; 95% CI, -0.6 to 0.0, P = 0.09).

Suzuki et al, 2015

• retrospective observational study
• 15,818 patients over a 12 year period in four Australian ICUs
• 64% of ICU patients received at least 1g paracetamol
• Patients who received paracetamol had a higher mean temperature and were more likely to have a pyrexia than those who did not receive paracetamol
• Patients who received any paracetamol had a lower in-hospital mortality (10% vs. 20%, P <0.001)
  • Multivariate logistic regression analysis found an independent association between use of paracetamol and a reduced in-hospital mortality (adjusted OR, 0.60; 95% CI, 0.53 to 0.68, P < 0.001) but this association was lost after adjustment for pyrexia.
• As with the HEAT trial there was an association between paracetamol and increased time to death

Young et al, 2012

• Retrospective observational study
• ~600,000 critically ill patients, comprising a derivation cohort from the ANZICS database and a validation cohort from the UK
• Baseline mortality was derived from those with a peak temperature of 36.5C – 36.9C
• Findings
  • From the ANZ cohort:
    • adjusted in-hospital mortality risk progressively decreased with increasing peak temperature in patients with infection
    • a peak temperature of 39C – 39.4C was associated with the lowest risk of death (adjusted OR, 0.56; 95% CI, 0.48 to 0.66)
    • In non-septic patients, the mortality risk increased with increasing temperature above 39.0C  (adjusted OR, 2.07 at 40.0°C or above; 95% CI, 1.68 to 2.55)
  • Similar findings were obtained from the UK cohort

Mullins et al, 2011

• 3-armed single-center RCT of n= 79 neurocritical care patients with T≥38C
  • 50% had intracranial haemorrhage
• intervention/ comparisons:
  • single dose of paracetamol 975 mg, a single dose of ibuprofen 800 mg, or a combination of both
• Oral temperatures were measured hourly for 6 h following medication administration.
• Outcomes:
  • Difference in area under the curve (AUC) for temperature reduction for ibuprofen plus paracetamol was -1.56C/h (P = 0.03) compared to paracetamol alone and -1.05C/h (P = 0.09) compared to ibuprofen alone
• Note that AUC is not the same as actual temperature reduction, in the combination group the difference in mean temepratures at t0 and t2-6 was only ~1.1C
• Lack of a placebo control makes it unclear how much of the treatment effects were due to regression to the mean
• Conclusion
  • Combination of ibuprofen and paracetamol may be more effective at decreasing temperature in neurocritical care patients, but any effect is small and clinical significance is questionable

Other studies

• Schortgen et al, 2012
  •  prospective observational study of 1,425 ICU patients with fever due to any cause
  •  increase in 28-day mortality (adjusted OR, 2.61; P = 0.028) with the use of NSAIDs a with the use of paracetamol (adjusted OR, 2.05, P = 0.01) in patients with sepsis
  • no difference in patients without sepsis
  • Physical cooling was not associated with a change in mortality, in either septic or non-septic patients
• Lee et al, 2012
  • observational study of critically ill patients
  • For non-septic patients with a maximum temperature of 38.5 °C – 39.4 °C the odds ratio for 28- day mortality was 5.13 (P < 0.007)
  • For a maximum temperature ≥ 39.5 °C the odds ratio was 13.4 (P < 0.001)
• Greenberg et al, 2010
  • observational study of critically ill patients with untreated fevers
  • paracetamol had little effect on the rate of cooling (mean 0.20C/h compared to 0.13C/h (95% CI of the difference, -0.1107 to -0.01204; P = 0.0152)
• Dippel et al, 2003
  • RCT of n=112 patients with acute ischaemic stroke
  • randomised to either 1g paracetamol or placebo q6h
  •  paracetamol group had a0.26 °C lower mean body temperature within 4 hours of treatment than those treated with placebo (95% CI, 0.07C – 0.46C)

AN APPROACH

• Paracetamol can be safely used in critically ill patients as an analgesic and anti-pyretic
• The balance between the beneficial and detrimental effects of paracetamol on body temperature may depend on the cause of fever, the severity of the fever and the nature of the critical illness
• There is a paucity of evidence for the efficacy of paracetamol as an analgesic in the critically ill patients
• Paracetamol has a limited antipyretic effect (~0.25C temperature reduction) in critically ill patients

References and links

LITFL

• Fever: Friend or Foe? (by Paul Young)
• CCC — Paracetamol
• CCC — Fever in Critical Illness (pending)

Journal articles

• Dippel DW, van Breda EJ, van der Worp HB, et al. Timing of the effect of acetaminophen on body temperature in patients with acute ischemic stroke. Neurology. 61(5):677-9. 2003. [pubmed]
• Greenberg RS, Chen H, Hasday JD. Acetaminophen has limited antipyretic activity in critically ill patients. Journal of critical care. 25(2):363.e1-7. 2010. [pubmed]
• Jefferies S, Saxena M, Young P. Paracetamol in critical illness: a review. Critical care and Resuscitation. 14(1):74-80. 2012. [pubmed] [free fulltext]
• Karamanou M, Liappas I, Antoniou Ch, Androutsos G, Lykouras E. Julius Wagner-Jauregg (1857-1940): Introducing fever therapy in the treatment of neurosyphilis. Psychiatriki. 24(3):208-12. 2013. [pubmed]
• Lee BH, Inui D, Suh GY, et al. Association of body temperature and antipyretic treatments with mortality of critically ill patients with and without sepsis: multi-centered prospective observational study. Critical care (London, England). 16(1):R33. 2012. [pubmed]
• Mullins ME, Empey M, Jaramillo D, Sosa S, Human T, Diringer MN. A prospective randomized study to evaluate the antipyretic effect of the combination of acetaminophen and ibuprofen in neurological ICU patients. Neurocritical care. 15(3):375-8. 2011. [pubmed]
• Saxena MK, Taylor C, Billot L, et al. The Effect of Paracetamol on Core Body Temperature in Acute Traumatic Brain Injury: A Randomised, Controlled Clinical Trial. PloS one. 10(12):e0144740. 2015. [pubmed] (aka PARITY trial)
• Schortgen F, Clabault K, Katsahian S. Fever control using external cooling in septic shock: a randomized controlled trial. American journal of respiratory and critical care medicine. 185(10):1088-95. 2012. [pubmed]
• Suzuki S, Eastwood GM, Bailey M. Paracetamol therapy and outcome of critically ill patients: a multicenter retrospective observational study. Critical care (London, England). 19:162. 2015. [pubmed]
• Young PJ, Saxena M. Fever management in intensive care patients with infections. Critical care (London, England). 18(2):206. 2014. [pubmed]
• Young PJ, Saxena M, Beasley R, et al. Early peak temperature and mortality in critically ill patients with or without infection. Intensive care medicine. 2012. [pubmed]
• Young P, Saxena M, Bellomo R, et al. Acetaminophen for Fever in Critically Ill Patients with Suspected Infection. The New England journal of medicine. 373(23):2215-24. 2015. [pubmed] (aka HEAT trial)