Letters NICE on suspected sepsis

Authors’ reply to Eisenhut

BMJ 2016; 355 doi: https://doi.org/10.1136/bmj.i5787 (Published 02 November 2016) Cite this as: BMJ 2016;355:i5787
  1. Saul Faust, professor of paediatric immunology and infectious diseases and director1,
  2. Norma O’Flynn, clinical director2,
  3. Caroline Keir, guideline commissioning manager3,
  4. Martin Allaby, consultant clinical adviser3
  1. 1NIHR Wellcome Trust Clinical Research Facility, Faculty of Medicine, University of Southampton and University Hospital Southampton NHS Foundation Trust, Southampton, UK
  2. 2National Guideline Centre, Royal College of Physicians, London NW1 4LE, UK
  3. 3Centre for Guidelines, National Institute for Health and Care Excellence, London SW1A 2BU, UK
  1. s.faust{at}soton.ac.uk

NICE guideline 51 covers the identification, risk assessment, and immediate management of people with suspected sepsis.1 It was not commissioned to replace other NICE guidance of specific conditions or to be a textbook of infectious diseases. In the absence of definitive biomarkers to diagnose sepsis, and in the face of many non-specific symptoms, the NICE guideline provides a pragmatic approach, recognising that at any stage other conditions may be diagnosed and more appropriate management pathways followed.

Eisenhut describes three scenarios.2 Firstly, a specific infection (such as encephalitis) that would not necessarily be recognised as sepsis should be treated according to the best care pathway for that condition. Any child or adult with symptoms or signs suggestive of encephalitis, but not meeting any high risk sepsis criteria, should be managed with appropriate investigations and treatments. For children under 5, the NICE feverish illness guideline3 can be followed if high risk sepsis is not identified. If encephalitis is strongly suspected in children of any age, the national best practice consensus guideline that Eisenhut cites should be used.4

Secondly, lactate was not found by the NICE group or others5 to correlate strongly to risk of mortality. This is why the guideline uses lactate to guide the immediacy of fluid therapy and further referral rather than as an independent risk factor. Eisenhut is therefore incorrect to say that more than 30% of patients with bacteraemia could be missed using a cut-off of 2 mmol/L of lactate, as the NICE guideline does not use lactate as part of the risk assessment.

Finally, the NICE guideline clearly states that immediate antibiotics should be given if one high risk criterion is present alongside the clinical suspicion of sepsis. This includes cases of urinary tract infection, or any suspected infection, where poor peripheral perfusion (evidenced by parameters including heart rate or respiratory rate above the thresholds outlined in the NICE guidance) indicates a high risk of mortality due to sepsis. Rigors were not considered by the NICE group to be an independent risk factor but are certainly part of the assessment leading to a suspicion of infection. To avoid overtreatment with antibiotics, the guideline allows de-escalation to specific care pathways for identified specific infections. However, non-specific symptoms in the context of one definite high risk criterion and no alternative diagnosis should certainly lead to immediate antibiotic therapy.



View Abstract

Log in

Log in through your institution


* For online subscription