A child with a fever

Author: Frances Copp / Reviewer: Lucine Nahabedian / Codes: / Published: 12/09/2018 / Reviewed: 23/04/2024

Febrile children compete for the most common non-traumatic paediatric presentation in the Emergency Department (ED), causing concern for parents worldwide. Your mission: to establish if the child is clinically stable and to risk stratify them for serious bacterial infection. This guide, by no means exhaustive, aims to talk you through a basic approach to assessment and management, with a reminder of those causes you really don’t want to miss.

Key points:

  • Source is key. If you find a source, further investigation may not be necessary
  • Look for signs of serious illness and suspect the ‘scary stuff’ in all children – that way you are less likely to be caught out
  • Investigate as indicated, not as a routine
  • Antibiotics are not always the answer
  • If in doubt, a longer period of observation and a senior opinion never go amiss.


Before even calling the patient’s name, take a moment to observe the child in the waiting room: How do they look? Charging around the play area with toy in hand? Subdued, super-glued to their guardian? Distressed? Unconscious? Posturing?

Whilst certainly not the only factor to be taken into account (see NICE guidelines NG143 (1)), much can be gleaned from a childs level of activity and interaction with those around them particularly when trying to determine whether you are dealing with a septic child or yet another happy URTI. However, beware: younger infants may not respond as dramatically as older children to illness. In babies, particularly under 3 months of age, they may only have non-specific symptoms e.g. reduced feeding and level of alertness. They also do not always mount a fever.

First impressions can also be used as a reference point: if the child presents as grizzly, clingy and refusing fluids/food, with a fever and raised heart rate on initial observations, but an hour or two after administration of antipyretics you see before you a bouncing, interactive, feeding child with normal heart rate and settling temperature, you and the parents may be reassured. This may even be sufficient reassurance for discharge.

It is also worth remembering that first impressions go both ways. Approaching with a smile, getting quickly to their level (height and humour) and plying younger children with bubbles and distraction toys goes a long way to help you fully assess a child’s level of distress and gain trust from parents.

Frequently, parents exhibit ‘fever phobia‘: they are often concerned by the fever itself, the perceived degree of fever, regularity of fevers and the potential for fevers to continue to spiral with dangerous outcome. If specific concerns are uncovered early in the discussion – especially with any related parental experience – you may be guided towards any targeted reassurance and many of these fears may be abated.

Firstly, fever usually represents a natural physiological response to infection and damage is rarely a direct result of the fever itself. Whilst fever plays its part in the fight against invading pathogens, the comfort brought by reducing the temperature has long been thought to outweigh the benefit of letting the fever run its course as we all know, when febrile with an infection on board, you feel dreadful. Therefore, current advice is to dose-up, if there is evidence of distress.

Figure 1: Use of antipyretics, taken from NICE Guidelines: ‘Fever under 5s’

Complications and adverse outcomes are usually related to the underlying aetiology and associated fluid loss. The specific value of the fever is usually irrelevant: higher temperatures (40-42 degrees Celsius) are not linked to tissue damage nor to the development of febrile seizures. (4) It might be helpful here to note the confusion around the disputed value of temperature regarded as fever interestingly, the NICE Guidelines only mentions the figure of 38 degrees Celsius and only in association with infants under 3 months. Clinicians dealing with fever in children will often take this value to be the starting point for a fever, and any value 37.5-38 classed as low grade. Due to hypothalamic control, temperatures caused by infection rarely exceed 42 degrees.

It is useful to delve into the history of the fever:

  • Fever onset and duration: clarify if there are daily high temperatures and if there has been any break in the fever (and if so, for how long particularly important in presentations of over 5 days of fever).
  • Any recorded temperatures at home (if normothermic on presentation). Occasionally parents will have a different idea of ‘fever’ to those clinically recognised.
  • Associated febrile symptoms: behaviour change with fever (lethargy, off feeds/not drinking)? Improving when fever settles? Rigors? Febrile seizures?
  • Anti-pyretic use and response: regular use and, if so, paracetamol, ibuprofen or both, and at what dose? Always note the time of the last dose(s) given!
  • Any focal infective symptoms?
  • Rash? If so timing in relation to fever, where did it start and any spread?
  • Oral intake and urine output – dehydration?
  • Vaccination history
  • Birth history particularly relevant in younger babies: Preterm? Perinatal complications such as postnatal/intrauterine infection?
  • Past medical history: any recurring infections? Immunocompromise? (usually volunteered early by parents)

Suspicion of more severe, or more specific, aetiology may be raised if any of the following are revealed:

  • Recent travel abroad, or recent contact with travellers
  • Contact with unwell individuals
  • In the case of GI upset – suspicious food history and associated time frame: restaurants? Take-aways? BBQ? Others unwell with same symptoms from same event? Meat/fish?

It is also worth considering in the social history others who may be affected for example, any contact with immunocompromised individuals, pregnant women or young babies?

As with any good history, you will often be able to predict the source of the fever even if the child is unable to localise any symptoms. Be prepared to think laterally for example, if a child is refusing food and fluids could the source of fever be visible in the mouth or pharynx?


The younger the child, the more important the examination. Take note of the NICE table for signs of serious illness and bear in mind the Paediatric Sepsis 6 criteria will be age dependent due to the differences in the parameters for their observations. The UK Sepsis trust have tools (figure 2) to aid identification and management of sepsis in under 5, 5-11 and over 12s (4) which are endorsed by NICE.

Figure 2: Paediatric Sepsis 6 identification in under 5 years.

Don’t be afraid to expose fully to check for a rash and any inflammation around the genitalia. Wiggle joints for tone and tenderness and never, ever, forget ENT and lymph nodes! No child looks forward to a good tonsillar inspection and whilst this may be challenging, recruiting the parents and using distraction techniques will help you get the best view possible. Position is paramount. It’s worth leaving to the end of your examination, as you will probably make them cry! On your way in, make sure you take note of the appearance of the tongue and oral mucosa. The only exception to this would be if you have any suspicion of epiglottitis/severe croup in which case, further aggravation is best avoided

Be careful when looking for a rash: petechiae may be very subtle to begin with. Clarify the cause of any bruises with the child or parents.

It is useful to remember a non-infective cause you do not want to miss: Kawasaki disease. Principally affecting young children (<5yrs), this systemic vasculitis can result in cardiac disease (coronary aneurysms), affecting 20-25% of those left untreated. (6) Any sign or suspicion should prompt urgent discussion and review with Paediatric seniors.

Figure 3: Table of Signs of Kawasaki Disease, adapted from NICE Guidelines ‘Fever in under 5s’

Pyrexia of unknown origin (PUO)

Consider pyrexia of unknown origin in any child with fever for over 7 days without clear source.

Consider referral to rheumatological and infectious disease specialists but remember that a source cannot be found in up to 30% of cases. (7)


Whilst most parents will be understandably wary of lumbar punctures, even if clinically indicated (see NICE NG 143), blood tests are a mythicized phenomenon for many. With explanation of the limitations of the tests available (CRP and WCC alone are not reliable markers of severity of infection), explanation of the potential distress caused by cannulation, and reassurance that management would not be changed by blood test results, any request for blood tests are usually retracted. That being said, it is always reasonable to check a bedside BM if lethargic or vomiting, particularly with a history of poor oral intake.

There are instances, however, in which blood tests are beneficial – principally FBC, CRP, blood culture and lactate are indicated in the following febrile circumstances:

  • Infant under 3 months of age: due to maternal immunoglobulin cover and their own immature immune system, presentation with fever is more suspicious of bacterial sepsis (Beware that those under 1 month are likely to have a full septic screen and those between 1-3 months can have partial septic screens. The evidence behind this is currently being evaluated along with the terminology we use for this process (8))
  • Any suspicion of meningococcal infection or meningitis: petechial/purpuric rash, fever, irritability/lethargy, photophobia, bulging fontanelle any of these should raise your suspicions, but meningitis should be considered as rule-out differential. Remember to take a clotting sample and meningococcal PCR!
  • Suspected PUO
  • Any child over 3 months of age with fever of unknown source and 1 or more amber or red feature(s) from the traffic light table
  • If there is any concern that the child may be in shock.

Specific rapid testing may also be indicated depending on signs and symptoms e.g. monospot for EBV, malaria testing, or ASOT and throat swab if there are concerns over the possibility of streptococcal infection (also useful to take a viral throat swab at the same time!)

For the vast majority, less-invasive investigations will suffice: a clean-catch sample of urine for bedside dipstick testing, if indicated. A stool sample may be collected if relevant (always request if any evidence of bloody diarrhoea).

X-rays are more contentious. If you can hear unilateral crepitations and the child is relatively well, X-ray findings are unlikely to change your management plan of discharge with oral antibiotics. (BTS guidelines (9)).


In most cases, a fever will be the result of a virus with clear localising signs. Rarely, other demons are at large. However, it is important to acknowledge that ‘just a virus’ may cause an otherwise well child to decompensate and therefore it is always necessary to be on the lookout for signs of compromise.

If you have found a source and feel your patient is well enough to head home, you may now be facing the antibiotic conundrum: to give or not to give? For this, bear in mind the question: ‘what am I treating and for whom?’ do not be swayed by parents if you feel antibiotics are not indicated! Stressing the relatively high risk of complications and likelihood of minimal impact on the course of illness is usually enough to dissuade any pro-antibiotic parents. The following poster from Edward Snelsons gppaedstips blog site may also be of aid:

Figure 4 (10)

For many URTIs it is very difficult to distinguish between viral and bacterial causes however, if there is a possibility of secondary complication due to untreated bacterial infection (for example, streptococcal infection and Scarlet fever (See RCEM Learning module (11)): antibiotic cover should be carefully considered. Scoring systems can help towards this (for example CENTOR score for likelihood of Streptococcal tonsillitis), although these also have their limits. The best approach is to treat the patient in terms of risk-benefit: if a child has severe or prolonged symptoms, complications, typical presentation for a particular antibiotic-susceptible pathogen, the reported 10% risk of complication from taking antibiotics may be outweighed by the benefits of their use.


Should you decide that your patient is fit to go home, now is the time to further empower their parents in the management of fever. Explain what to expect in terms of symptoms, give advice regarding the use of anti-pyretics for comfort, advise regular fluids, and provide strong and specific safety netting for signs of worsening infection and dehydration. Remember, what you see now may not represent the patient later in the course of infection if any deterioration, parents must be encouraged to seek help.


  1. National Institute for Health and Care Excellence (NICE). Fever in under 5s: assessment and initial management. NICE guideline [NG143] 2019. Last updated: 26 November 2021.
  2. Schmitt BD. Fever phobia: misconceptions of parents about fevers. Am J Dis Child 1980; 134: 176-181.
  3. El-Radhi AS. Fever management: Evidence vs current practice. World J Clin Pediatr. 2012 Dec 8;1(4):29-33.
  4. National Institute for Health and Care Excellence (NICE). Suspected sepsis: recognition, diagnosis and early management. NICE guideline [NG51] 2016. Last updated: 19 March 2024.
  5. The sepsis trust Clinical Tools
  6. BMJ Best Practice – Kawasaki disease. Last reviewed: 2024.
  7. Fox SM. Fever of Unknown Origin. Pediatric EM Morsels, 2015.
  8. Roland D, Munro A. Time for paediatrics to screen out sepsis screening BMJ 2023; 381 :p1327.
  9. Harris M, Clark J, Coote N, et al. British Thoracic Society guidelines for the management of community acquired pneumonia in children: update 2011. Thorax 2011;66:ii1-ii23.
  10. Paediatrics for Primary Care (and anyone else) – Decision Fatigue and What to Do About It – When to Use Antibiotics for URTI, AOM and Tonsillitis in Children. 2018.
  11. Abela N. GAS, iGAS and Scarlet fever. RCEMLearning, 2022.

Leave a Reply