Fever: Fear and Tradition

2 December, 2010 by: colinparker

Post to Twitter

“Doc, he’s burnin’ up!”  Fever is probably the commonest presenting complaint we manage in Paediatric Emergency Departments.  The cause is usually an infection… but not always.  The infection is usually a benign, self-limiting viral illness… but not always.

Join us for this podcast as we tease out the facts from the fiction, the myths and the mystery.

To listen, just click on the ‘play’ triangle below, or you can subscribe via your podcatching software (such as iTunes).

Fever Basics (Fear and Tradition)

CP/all: welcome, disclaimer, hello, intro


CP: Physiology / theoretical survival advantage of fever

SF: Methods of measuring: core (rectal), tympanic (not under 6 months), oral, axillary, ‘forehead strips’, ‘feels hot to parents’
KB: Definition of a fever, significant ‘cutoff’ values eg depending on age (neonate, 1-3months, 3-24 or 36 months)

Causes of fever:

KB: infections (viral, bacterial, rickettsia, malaria, others ) central theme = benign viral vs serious bacterial
SF: haematological/ oncological (lymphoma, leukaemia, Wilms, Neuroblastoma, others)
CP: auto-immune/ chronic inflammation (JIA, SLE, etc)

Periodic fevers:

See:  Periodic Fever Syndrome
and Causes of Cyclical Fever in Children

KB: Familial Mediterranean Fever
SF: Cyclical Neutropaenia
CP: Hyper-IgD syndrome
KB: TNF-Receptor Associated Periodic Syndrome (TRAPS)
SF: PFAPA syndrome

CP: drug-induced fevers
KB: factitious / induced illness
SF: don’t forget Kawasaki Disease
CP: idiopathic fever?

CP: FWS vs PUO (2 weeks)

Risk Stratifying FWS (SBI vs benign viral illness)

CP: past strategies – risk minimisers vs test minimisers, use of WCC
Changing landscape post-Pneumococcal Conjugate Vaccine (following the US experience)
“Needle in a haystack” problem and the Ian Everitt corollary…

SF: factors to consider in risk-stratifying:

  • age
  • height of fever (or not)?
  • clinical findings / source (incl “soft”/ co-existent signs like slightly red throat, pink TMs from fever itself)
  • urine sampling in those without clear clinical focus

KB: “well” vs “unwell” – hard to define, hard to teach!

Can we forget about “Occult Bacteraemia” now?

Treatment of febrile illness

SF: treat underlying infection:

  • clear source/focus: treat appropriately based on diagnosis and severity
  • no focus but unwell: screen (LP, CXR, Urine, BC) admit, IV AB’s pending negative cultures
  • well but FWS: follow-up strategy 12-24 hrs GP/ED, (+/- IM antibiotics, Blood cultures)- evolving

CP: supportive care: hydration, nutrition, observation, comfort

Antipyretics for comfort?

KB: arguments FOR antipyretics (feel better, look better, drink better, easier to assess clinically, placebo effect?)
SF: arguments AGAINST antipyretics (not natural – defence mechanism, medication side-effects – Reye Syndrome historically, ?wheeze from NSAIDs,  may prolong illness)

CP: physical cooling: methods (undressing, fan, tepid sponging, cool bath, “hydrotherapy”) benefits & risks

Fever Myths

CP: “Fever is Dangerous” (boiled brain)
SF: “Antipyretics prevent Febrile Convulsions”
KB: “Favourable Response to Antipyretics excludes Serious Bacterial Illness”
CP: “Social smile excludes Serious Illness”

Bass JW, Wittler RR, Weisse ME. Social smile and occult bacteremia. Pediatr Infect Dis J. 1996;15(6):541.
PubMed PMID: 8783353.

Advice to Parents on Discharge

CP: fever in perspective, supportive care, follow-up if necessary
All: specific reasons to return

CP/All: Summary

Goodbye for now Folks!
Next time, we will discuss: NICE Guideline CG47 – Feverish Illness in Children.
As always, we welcome your intelligent and insightful comments!

Post to Twitter



3 Responses to “Fever: Fear and Tradition”
  1. colinparker says:

    This is a long one… but probably worth setting aside an hour for, given how frequently we encounter febrile children in our work!

  2. Great site. A lot of useful information here. I’m sending it to some friends!

  3. Colin says:

    Thanks to Uncle Frank, who sent me this recent article looking at the utility of blood tests in Fever Without Source, in a population of 1-36 month old children immunised with Pneumococcal Conjugate Vaccine:

    Markers for bacterial infection in children with fever without source
    Sergio Manzano, Benoit Bailey, Alain Gervaix, Jocelyne Cousineau, Edgar Delvin, Jean-Bernard Girodias
    Archives of Disease in Childhood (advanced online publication 29 January 2011) – no PubMed listing yet!

    Their results suggest that blood tests are superior to clinical evaluation by a Paediatric Emergency Physician… but I have some concerns:

    Firstly, the clinical evaluation did not include a urinalysis or urine microscopy – and the commonest SBI (by far) was UTI (48 out of 54 Serious Bacterial Illness patients, 89%). Perhaps we could have avoided blood tests in these children.

    Secondly, more than 20% of the study population were excluded from analysis due to insufficient blood sample quantities – 93 out of 457 patients. Seven percent of the excluded patients had SBI (all were UTIs). This reminds me of the awkward situation where you decide to do blood tests, but are unable to get the required samples – do you bail out and revert to clinical observation, or keep trying repeatedly to get the blood sample?

    Thirdly, despite the authors minimising the importance of clinical evaluation, their results show (in Table 5 and Table 6) that a strong clinical suspicion of SBI (as evidenced by a Visual Analogue Scale of greater than 50%) increased the likelihood of SBI two-fold (LR+ 2.2) overall, and seven-fold (LR 7.5) when UTIs were excluded. This means that the child’s risk of SBI goes from 3% to 19%, when you exclude the UTIs, and the clinician is worried.
    I do accept that clinical evaluation is less helpful at excluding SBI though, with negative Likelihood Ratios (LR-) of 0.7 when the clinical suspicion of SBI was less than 25%. This takes the risk of SBI from 3% to 2% when UTI patients are excluded.

    This is a good prospective effort at figuring out how useful these blood tests are at predicting SBI in the FWS situation, but it would have been great if these tests were included as part of a clinical strategy including urine microscopy before blood tests, in all except the sickest and youngest patients. The urine microscopy may have missed a small number of UTIs, but I suspect that if one of the tests examined in this cohort of patients was “urgent microscopy of catheter specimen urine “, it would have out-performed all the other tests, including clinical evaluation.

    I worry that the pendulum is going to swing back towards doing more blood tests as a Risk Minimiser approach, if we accept these findings at face value.

    What do you think?


Leave a Reply