Febrile illness – Emergency management in children

Purpose

This document provides clinical guidance for all staff involved in the care and management of a child presenting to an Emergency Department (ED) with a febrile illness in Queensland.

This guideline aims to identify those infants and children at risk of serious bacterial infection or other significant illness who need timely treatment, while avoiding unnecessary investigations in the majority. The management of children with an unexplained fever for greater than one week or who have recently returned from overseas travel is beyond the scope of this guideline. For oncology patients refer to the Management of Fever in a Paediatric Oncology Patient Guideline [PDF 622.23 KB].

This guideline has been developed by senior ED clinicians and Paediatricians across Queensland, with input from Infectious Disease specialist staff, Queensland Children’s Hospital, Brisbane. It has been endorsed for use across Queensland by the Queensland Emergency Care of Children Working Group in partnership with the Queensland Emergency Department Strategic Advisory Panel and the Healthcare Improvement Unit, Clinical Excellence Queensland.

Introduction

Fever is one of the most common paediatric ED presentations. Identifying a focus of infection can be challenging especially in very young children. While most children fully recover, infection remains the leading cause of death in children aged less than five years.

Definition

Fever is defined as a temperature greater than or equal to 38°C.

Pyrexia of unknown origin (PUO) refers to any fever lasting 10 – 21 days without a cause identified on history, examination and basic investigations and is beyond the scope of this guideline.

Measurement

In neonates (age less than 28 days) temperature should be measured using an electric thermometer in the axilla. For children over 6 months an infra-red tympanic thermometer may also be used if it can adequately enter the external auditory canal.

Current evidence suggests that forehead infra-red thermometers are not yet accurate enough for use in clinical settings.²⁰ Forehead chemical thermometers are unreliable and should not be used.

A parent’s touch has been shown to have high sensitivity and low specificity for identifying a fever, however parental concern should be taken seriously.

Pathophysiology

Fever is a physiological response most often caused by an infective process, when exogenous pyrogens induce endogenous pyrogens, resulting in an elevated body temperature. The thermoregulatory centre then raises and maintains the body temperature to the new set point. This gives most children a degree of malaise and may negatively stress children with pre-existing cardiac, respiratory or neurological diseases. Fever is a generally beneficial adaptive response that promotes the immune response and inhibits the invading pathogen, potentially reducing the duration of certain infections.

Aetiology

In most children less than five years of age, fever is caused by a viral infection.¹

Less common causes include serious bacterial infection (SBI) such as urinary tract infection (UTI), pneumonia, bacteraemia, meningitis, or bone and joint infections, or conditions such as Kawasaki disease, vaccination reactions, arthritis, connective tissue disorders, malignancies, drug fever, inflammatory bowel disease and environmental causes. The most common SBI in children is a UTI followed by pneumonia.

Post-vaccination fever is common with a typical onset within 24 hours of immunisation and duration of two to three days.

Teething does not cause fever greater than 38°C.

Assessment

The aim of the assessment (history and clinical examination) is to identify children who:

• have a focus of infection (to enable appropriate investigations and, if needed, treatment)
• do not have an infective focus but require further investigation

History

Questioning should include specific information on:

• immunisation history
• immunosuppression (either by medical condition or treatment)
• history of fever and use of analgesia
• current or recent antibiotic use
• recent overseas travel

This guideline also does not consider the approach to fever in the returned traveller – refer to Assessing fever in the returned traveller²

Age

Febrile infants aged less than three months have a higher risk of SBI, with the risk greatest in the neonatal period. Young infants are more likely to present with nonspecific features (they lack the hypothalamic and immune system maturity to localise the infection) and can deteriorate rapidly. Some infants less than three months may not mount a fever in response to SBI, and hypothermia or temperature instability can also be signs of SBI.

In addition to the pathogens seen in older children, Group B Streptococcus, E. Coli, Herpes Simplex virus, Listeria monocytogenes, Salmonella, Enterovirus and Parechovirus infections are more common in neonates. Detecting other viral infections in children aged less than three months (most commonly RSV) lowers the risk of, but does not preclude, a SBI.³ The estimated incidence of a UTI amongst infants less than three months with laboratory-confirmed RSV infection ranges from 3.3 to 5%.^4,5

Children aged between three months and three years who have their immunity boosted with vaccinations are at a lower risk of SBI than younger children. In this age group, the presence of a recognisable viral syndrome (including bronchiolitis) predicts a very low incidence of bacteraemia or SBI.

Children over three years of age have mature immune systems so are at a lower risk of SBI. The ability of older children to verbalise symptoms assists in identifying a focus of infection.

Immunisations and immune status

The Haemophilus influenzae type b (Hib) and pneumococcal immunisations have dramatically reduced the risk of occult bacteraemia and SBI. Children are best protected when they have received the three dose course (given as part of the National Immunisation Program [PDF] at 2, 4 and 12 months for both and at 18 months for Hib fourth dose). The 13-valent conjugate pneumococcal (13vPCV) and Hib vaccinations have greater than 95% protection.

Children with congenital immune deficiency syndromes, sickle cell disease, HIV, asplenia, cancer, nephrotic syndrome, intracranial shunt, cochlear implant, immunosuppressive therapy or who are of Indigenous or Torres Strait Islander origin are at a greater risk for SBI, independent of vaccination status.

History of fever, prior antibiotic use and use of anti-pyretics

The height and duration of the fever and the response to antipyretics have failed to show any ability to differentiate severe from mild illness, or bacterial from viral infection.⁶

A diagnosis of Kawasaki disease should be considered in children with a fever of five or more days without a clear clinical source.

Antibiotic therapy prior to presentation can mask the signs and symptoms of a bacterial illness.

Examination

The examination should identify a source for the fever if possible, and specifically assess for any signs of toxicity or early markers of the possibility of SBI.

Pay attention to concerns expressed by the caregiver, particularly any reported changes in usual behaviour.

Seek urgent senior emergency/paediatric advice as per local practice for a child with fever and any of:

• toxic features (in child of any age)
• age less than 29 days
• age 29 days to 3 months without typical respiratory illness

Investigations

Refer to the relevant guidelines if the following conditions are suspected:

• Sepsis
• Meningitis
• UTI
• Septic arthritis
• Bronchiolitis
• Limp
• Community acquired pneumonia
• Periorbital cellulitis
• AOM

Age three months or more

Most children aged three months or older who are fully immunised, have no comorbidities and appear well do not require extensive investigation. Refer to the flowchart (child 3 months and above) [PDF 411.85 KB] for the approach to investigations in these children:

A child with PUO may require more specialised investigations that are not included here.

Age less than three months

Refer to the flowchart (child less than 3 months) [PDF 424.19 KB] for the approach to investigations in febrile infants aged less than three months:

Due to the higher risk of SBI and the challenges in reliable clinical assessment of toxicity in neonates (corrected age less than 29 days) a consistent approach to investigation is recommended, irrespective of clinical appearance.

Owing to the lower risk of SBI, infants aged 29 days to 3 months can be managed in a step-by-step approach with sequential evaluation of general appearance, urinalysis, and results of bloodwork.⁷ With careful clinical assessment, this approach allows identification of a group of infants at low risk of SBI who can be safely managed as outpatients without requiring a lumbar puncture or empiric antibiotics.

Additional investigations (e.g. serum electrolytes and glucose) may be required based on the clinical presentation.

Presence of an identified viral pathogen and subsequent risk of SBI

Whilst a positive result for a known viral pathogen may be reassuring to explain a child’s clinical symptoms, care must be taken to still make a thorough assessment to assess risk of a co-existent SBI.  In older children where risk stratification is easier, a positive viral test may be enough to avoid any further investigations, but in the <3m cohort that decision should involve a senior clinician.  Studies show that well appearing infants < 60 days with a positive viral test are significantly less likely (OR 0.23⁾ to have an SBI, however the risk remains ~ 3.8% (2.8% UTI, 0.8% bacteraemia and 0.2% meningitis) so is not negligible, and hence the need for senior clinician involvement for informed decision making ^{18, 19}.

COVID and SBI risk

Data on the risk of concurrent SBI in COVID positive children is scarce.  As with other viral pathogens, in the older child, assessment of this risk may be able to be made clinically, whilst in infants this assessment is likely to be less reliable.  A single study comparing rates of SBI in COVID positive infants vs age matched COVID negative febrile infants showed a similar relative risk ratio (0.22) to that reported for other viral infections.¹⁶ However in this single study the absolute risk of SBI in the both groups was significantly higher (8% vs 34%) than routinely reports rates, likely reflecting the high acuity presentations seen in hospitals during the pandemic.¹⁷ As such, senior clinician involvement should be sought for all febrile COVID + infants. Refer to CHQ-GDL-63327 – The Management of Children with COVID-19 (QCH) (health.qld.gov.au) for more information.

Full blood count and risk of SBI

Despite being widely used, there is little evidence to support the use of a FBC in the risk stratification for well appearing, immunised children.¹⁰

A systematic review found WCC of no value in ruling out a SBI in immunised children and less valuable than CRP for ruling in SBI.¹¹ A prospective cohort study found that total WCC and absolute neutrophil count were not sufficiently accurate to be used as screening tests for febrile children with possible SBI.¹² For infants less than 60 days, no parameters on the FBC have been found to accurately predict the risk of a SBI.¹³

Meningococcal, salmonella & staphylococcal bacteraemia’s do not typically elevate the WCC, whilst toxic shock, streptococcal and viral infections can commonly cause a lymphopenia. As does PIMS-TS.

C reactive protein and risk of SBI

CRP is an acute phase reactant and concentrations start to rise four to six hours after the onset of inflammation and peak around 36 – 50 hours. CRP is better than the FBC for detecting SBI, especially if used after 12 hours of fever, however establishing a level of CRP which can reliably determine low risk of SBI is challenging. .⁷ In the absence of a definitive cut off value, CRP should only be used in conjunction with clinical assessment or on advice from senior clinicians and according to local practice.

Management

Refer to the flowcharts for a summary of the recommended emergency management of febrile children less than three months [PDF 424.19 KB] and greater than or equal to three months [PDF 411.85 KB]. Management is based on the risk of SBI. See Paediatric Antibiocard - Empirical Antibiotic Guidelines [PDF 628.53 KB] and QLD Paediatric Sepsis Guidelines page for further information on antimicrobial treatment.

Supportive management of fever

Remove excess layers of clothing from the child. Over-enthusiastic physical cooling can be counterproductive by stimulating shivering and other heat-retaining reflexes. Oral fluids, if tolerated, should be encouraged to maintain hydration.

Antipyretics / Analgesics

Analgesics may be prescribed for an awake child to provide relief from discomfort caused by the fever or the underlying cause of the fever. Educate parents regarding fever (immune system’s response to infection) and role of analgesics (do not treat or shorten illness and may not).

Avoid aspirin as the uncommon possibility of Reye’s syndrome increases with varicella or influenza-like illnesses. There is also some evidence to counsel caution when using ibuprofen for infections related to Varicella and Group A streptococcus.¹⁵

There is some evidence to suggest that Ibuprofen reduces fever and discomfort faster than Paracetamol.¹⁴ While the popular dual therapy dosing regimens reduce the time with fever compared to monotherapy, there is no significant difference in resolution of discomfort.¹⁶ Alert parents to the safety concerns that have been raised over recommending two drugs with different dosing regimens for little gain.¹

Management of fever with petechial rash

Seek senior emergency/paediatric advice as per local practice for unwell child with purpuric rash not consistent with Henoch-Schonlein purpura.

Refer to the Guideline for Petechiae and Purpura: Emergency Management in Children [PDF 365.08 KB].

Escalation and advice outside of ED

Clinicians can contact the services below to escalate the care of a paediatric patient as per local practices. Transfer is recommended if the child requires a higher level of care.

Refer to the guideline for sepsis, meningitis, UTI, septic arthritis and bronchiolitis as indicated.

Critically unwell or rapidly deteriorating child

Non-critical child

Inter-hospital transfers

Disposition

When to consider discharge from ED

Consider discharge ONLY on senior emergency/paediatric advice for:

• all febrile children aged 29 days to 3 months
• children aged 3 months or more with features of SBI who are not fully immunised

Consider discharge for children who meet ALL of the following criteria:

• no toxic features
• no other investigations or IV treatment required
• no features of SBI
• able to maintain adequate oral intake to maintain hydration
• can be safely managed at home and return in the event of deterioration (consider time of day, parent/carer comprehension and compliance, access to transport and distance to local hospital)

On discharge, provide parent/caregiver with a Fever fact sheet and advice to seek medical attention earlier if symptoms worsen or they have other concerns about their child’ health prior to their scheduled appointment.

Follow-up

• with GP within 24 – 48 hours to check progress and receive outstanding test results.

Follow-up must be arranged prior to discharge for children aged 29 days to 3 months.

When to consider admission

Admission is required for the following febrile children:

• suspected sepsis
• age less than 28 days
• need for ongoing management

Consider admission for the following children:

• age 29 days to 3 months (even if appear well may require longer period of observation)
• features suggestive of SBI
• inability to maintain adequate oral intake to maintain hydration
• unplanned return within 24 hours of initial assessment
• parents/caregivers unable to safely care for child at home and return in the event of deterioration

Related documents

Guidelines

• Sepsis
• Meningitis
• Management of Fever in a Paediatric Oncology Patient
• Urinary Tract Infection
• Assessing fever in the returned traveller
• Petechiae and Purpura: Emergency Management in Children

Health A to Z fact sheets

• Fever in children
• How to collect a clean urine specimen