Have you ever seen a child in your emergency department with a fever – he asks sarcastically? At the ginormous community hospital where I work, we see about 25,000 kids each year in our ED and about half of them present with fever. Yes, there still exists fever phobia in our society, which brings hoards of worried parents into the ED with their febrile kids. For most of these kids it’s relatively straight forward: Most kids with fever have clinical evidence of an identifiable source of infection – a viral respiratory infection, acute otitis media, gastro, or a viral exanthem. However, about 20% have Fever Without a Source despite your thorough history and physical exam.
A small but significant number of this 20% without an identifiable source of fever will have an occult bacterial infection – UTI, bacteremia, pneumonia, or even the dreaded early bacterial meningitis. These are all defined as Serious Bacterial Infections (SBI), with occult UTI being the most common SBI especially in children under the age of 2 years.
In the old days we used to do a full septic work-up including LP for all infants under the age of 3 months, but thankfully, times have changed in the post-Hib and pneumoccocal vaccine age, and we aren’t quite so aggressive any more with our work-ups. Nonetheless, it’s still controversial as to which kids need a full septic workup, which kids need a partial septic workup, which kids need just a urine dip and which kids need little except to reassure the parents.
In this episode, with the help of Dr. Sarah Reid and Dr. Gina Neto from the Children’s Hospital of Eastern Ontario, we will elucidate how to deal with fever phobia, when a rectal temp is necessary, how to pick out the kids with fever that we need to worry about, how to work up kids with fever depending on their age, risk factors and clinical picture, who needs a urinalysis, who needs a CXR, who needs blood cultures and who needs an LP, and much more….
Sarah Reid’s Best Case Ever is here.
Fever in a child is a common emergency department presentation. About 20% will have fever without an identifiable source, and a small but significant number of these children will have an occult, serious bacterial infection (SBI) (1).
UTIs are the most common occult SBI (2), especially in children
By definition, fever is an oral temperature >38°C, or rectal temperature >38°C. Take a rectal temperature in toddlers, infants and neonates; axillary and tympanic temperatures are less accurate for core temperature.
Is Fever Dangerous & Do We Need to Treat it?
Fever itself is the natural body’s response to fighting infection, and does not in itself cause harm. However, fever causes dehydration; treat kids to help them feel better and stay hydrated.
The precise height of the fever is not as important as it’s duration in predicting bacteremia. Children with infection as a cause for their fever almost never mount a fever high enough to be dangerous (>41.5°C); these very high temperatures are typically only seen in non-infectious causes of hyperthermia.
The value of treating the fever in the ED besides patient comfort and minimizing dehydration, is so that the child can be re-examined when afebrile to help risk stratify for SBI and to counsel the parents. If the child’s vital signs and clinical picture continue to be worrisome after their fever has been corrected, then SBI should be suspected. However, if, after normalizing the temperature, the child has normal vital signs and is clinically ‘well’, then SBI is less likely and the parents can be reassured that fever itself is not dangerous.
Temperature Corrected Heart Rate and Respiratory Rate
Heart Rate increases by approx 10 beats/min and Respiratory Rate by 5 breaths/min for every Celsius degree (1.8 degree of Fahrenheit) of fever >38°C.
For example, if the temperature is 40°C and HR 144 -> subtract 2×10 -> corrected HR 124
If the child has tachycardia after being corrected for fever, consider other contributors (pain, crying, early compensated shock). Check perfusion, ask about urine output, and have a high degree of suspicion for dehydration and sepsis.
Acetaminophen vs Ibuprophen for Fever Reduction
Studies suggest that ibuprophen may be superior to acetominophen for treating pain and fever in children. Using both in combination of ibuprophen and acetominophen in an alternating pattern may be more effective, however the complex dosing associated with combination therapy increases potential for errors. A dosing handout can help – prompt parents to record times and dosages given.
Recommend no more than 3 doses of 10mg/kg of ibuprophen/24h and 4 doses of 15mg/kg of acetaminophen/24h, and emphasize keeping the child hydrated, to reduce risk of drug toxicity. Some experts suggest only giving antipyretics on a prn basis for fever, rather than on a schedule, so as to avoid potential complications with scheduled dosing (liver complications have been reported in children on q4h acetominophen, and renal complications have been reported in children on q6h ibuprophen). There is no good evidence that these medications prevent febrile seizure.
Is the Parent’s Touch Accurate at Predicting Fever?
The parent’s touch has been shown to be fairly accurate for identifying children with fever. Therefore a child without a measured fever at home but brought in for “tactile fever” should still receive the same history and physical assessment as the febrile child, even if they are afebrile in the ED.
Particularly important, is whether antipyretics were given, and the duration of fever. If the child is
An excellent review of the literature on treating pediatric fever on SGEM.
PEDIATRIC FEVER WITHOUT A SOURCE
Fever without a source:
Child <3 years old who after an initial history and physical does not have an identified cause of their fever. Approximaately 5% are likely to have a bacterial infection (usually a UTI).
As opposed to Fever of unknown origin:
Fever for 2 to 3 weeks, without a known cause after initial investigations. These fevers may be caused by more unusual illnesses (rheumatologic, oncologic). However, the cause is still usually infection, most often viral.
Key questions to ask in the history in children with fever without an apparent source
- Duration of fever
- Recent surgeries and underlying medical diseases
- Previous infections
- Immunization status (especially the first two doses, i.e. at 2 and 4 months, of Hib and Pneumococcal vaccinnes
Update 2019: A secondary analysis of the PECARN study of 4821 febrile infants (<60 days old) showed height of fever was associated with 3 major types of serious bacterial infection (SBI): for every 0.5 degree Celsius increase in temperature, the odds of SBI increased by 48%. Abstract
Physical examination for fever without a source
Watch behaviour and mental status; watch them walk (for septic arthritis/osteomyelitis); and examine the skin, joints, and abdomen very carefully.
5 Sources of Bacterial Infection to Look for in Fever Without A Source
1. Urine – UTI #1
2. Skin – cellulitis
3. Abdomen – appendicitis, abscess etc.
4. Joints – septic arthritis
Pediatric Urinary Tract Infections
What are the risk factors for UTI in children?
- History of previous UTI
- Temp >39°C, and without an apparent source, >24h
- Ill appearance
- Supra-pubic tenderness, (or dysuria/frequency/low-back pain or new-onset incontinence in children old enough)
- Females, uncircumcised boys and non-black race
Who to test (see JAMA guideline article 3):
< 3 months: check urine in all babies with fever without source
3-24 month: check all girls, and boys if >1 risk factor, or if circumcised and >2 risk factors
>24 months: check all girls, all symptomatic uncircumcised boys, and circumcised boy who had several symptoms suggesting UTI
Consider checking the urine even in children who have had another source for their illness identified; in children <60 days old with bronchiolitis, many will also have a UTI! (4)
How to get the urine?
obtain a sample by catheterization (and send every sample for culture, as the urinalysis may be normal in the presence of true infection)
>2 months of age to toilet trained: bag urine okay to screen by microscopy, however if it is positive (>10-20 white cells), a sterile culture needs to be obtained by catheter
Toilet-trained kids: obtain midstream urine.
Update 2017: A recent study by Chaudhari et al. in the Annals of Emergency Medicine has shown that the diagnostic accuracy of Lucocyte Esterase and pyuria depends on urine concentration, suggesting that specific gravity should be taken into consideration when interpreting a urinalysis in kids. Hence, it may be warranted to use a lower pyuria threshold in a dilute urine sample compared to a concentrated urine sample to diagnose a pediatric UTI. Abstract
Update 2017: A recent RCT has found a simple stimulation method for faster clean catch urine collection from infants in the ED. They compared the Quick-Wee method with gentle suprapubic cutaneous stimulation using gauze soaked in cold fluid (n=174) vs. standard clean catch urine method with no additional stimulation (n=170). The Quick-Wee method resulted in a significantly higher rate of voiding within five minutes compared with standard clean catch urine (31% v 12%, P<0.001). Ultimately, if you need an infant urinalysis, and clean catch is acceptable, the Quick Wee method can help you obtain a sample in less than 5 minutes. Abstract
Which children with UTI require admission?
Generally children <2 months of age should be admitted.
Well appearing children >2 months can usually be discharged home on antibiotics with good follow up, unless they are dehydrated or parents are unreliable.
Treatment of UTI
Antibiotic options depend on local antibiotic resistance patterns, however, in our experts’ catchment area,
In hospital: IV ampicillin and gentamicin
Oral: cephalexin for most, or cefixime for infants 2-6 months old, or those you are worried have a complicated UTI or urinary tract abnormalities.
Do all children with a first-time UTI need a renal ultrasound and voiding cystourethrography (VCUG)?
All need an ultrasound, and then only VCUG if there is hydronephrosis found on the ultrasound.
Which children need further investigations for Fever Without a Source?
Indications for CXR in Pediatric Fever Without a Source?
- Fever >5 days,
- or cough >10 days,
- if the temperature is persistently high (>40), or if a white count >20 with no other explanation
Remember: Examine patients carefully for “quiet tachnypnea” beyond what is expected for the height of fever.
Indications for Bloodwork and Blood Cultures in Pediatric Fever Without a Source?
A well-appearing, immunized child with a fever typically does not need blood work or cultures (6).
CRP and Pro-calcitonin might be helpful in risk-stratifying patients with fever without a source, it hasn’t become standard of practice or available in most centres.
Update 2017: This article validates the Step-by-Step approach to assessing febrile infants without a source, which has a very high negative predictive value for serious bacterial infection but should be used with caution in infants presenting within 2 hours of onset of fever and in infants 21-28 days of age. Abstract
Update 2019: In a prospective observational study of 913 infants (<60 days old), a clinical decision rule (PECARN) to identify low risk , febrile infants for serious bacterial infection (SBI) was developed when the following parameters were met: a negative urinalysis, ANC ≤4090/µL, and procalcitonin ≤1.71 ng/m. (NPV – 99.6%; NLR – 0.04). This rule has not been externally validated. Abstract
Guidelines by age for fever and the septic workup
Our experts recommend using chronological age (how many days since the child’s birth, rather than adjusted gestational age) to help guide decision making for septic workup. *However premature infants with a complex hospital course are already high-risk patients, so these babies usually require more thorough investigations.
A full septic workup including LP is recommended for babies under 28 days, because they have the highest risk of SBI. This includes routine blood work and culture, urinalysis and culture, and lumbar puncture (cell count, protein, glucose, culture, gram stain and culture, and viral studies).
Which infants should receive acyclovir for meningitis? If you suspect meningitis based on physical, or LP results (especially in <14 days), start acyclovir. HSV also causes pneumonitis and hepatitis, so also check for these if you are suspicious of HSV meningitis.
For infants 29 days to 90 days, use the low-risk criteria (see below). If they are well appearing, with no obvious source of infection, no complex past medical history, normal laboratory criteria (WBC count, normal urinalysis, and normal stool white count if diarrhea is present?), they can usually be sent home if they have reliable parents and good availability for follow up in 24h. These infants have a chance of SBI of about 1.5%, usually UTI, so make sure urine is sent also for culture.
American Academy of Pediatrics Fever Without a Source Guidelines
For a concise bedside summary of fever without a source in 29-90 day old infants see Michelle Lin’s ALiEM card.
Now Test Your Knowledge
Update 2016! Justin Morgenstern publishes a great analysis of the most important pediatric UTI literature relevant to EM to date on First10EM
For more on Paediatric Emergencies download our free interactice eBook EM Cases Digest Vol. 2 Pediatric Emergencies
Dr. Helman, Dr. Reid and Dr. Neto have no conflicts of interest to declare.
1) Baraff, L. Management of Infants and Young Children with Fever without Source. Pediatric Annals 37:10, October 2008. Free Full PDF
2) Subcommittee on Urinary Tract Infection, Steering Committee on Quality Improvement and Management Pediatrics. 2011;128:595-610.
3) Shaikh, N. Does This Child Have a Urinary Tract Infection? JAMA, December 26, 2007 Vol 298 No 24. Full Free PDF
4) Levine, D. Risk of Serious Bacterial Inection in Young Febrile Infants with Respiratory Syncytial Virus Infections. Pediatrics. 2004;113;1728. Full Text
5) Urinary Tract Infection: Clinical Practice Guideline for the Diagnosis and Management of the Initial UTI in Febrile Infants and Children 2 to 24 Months. Pediatrics 2011;128;595; originally published online August 28, 2011. Free Full PDF
6) Wilkinson, M. et al. Prevalence of occult bacteremia in children aged 3 to 36 months presenting to the emergency department with fever in the postpneumococcal conjugate vaccine era. Academic Emergency Medicine 2009; 16:220-225. Free Full PDF
For more EM Cases content on Pediatric Emergencies check out our free interactive eBook,
EM Cases Digest Vol. 2 Pediatric Emergencies here.
For more Pediatric EM learning visit one of EM Cases newest collaborators trekk.ca – Translating Emergency Knowledge for Kids (TREKK) is a growing network of researchers, clinicians, health consumers and national organizations who want to accelerate the speed at which the latest knowledge in children’s emergency care is put into practice in general EDs – rural, remote or urban.