Tim Sell. Fever and a limp, Don't Forget the Bubbles, 2014. Available at:
2 1/2 year old John is brought into the emergency department by his mother. He has been refusing to weight bear for four days.
We often see children with pyrexia and have to decide on whether or not they have a serious underlying bacterial infection. When the parent describes, or uses the word ‘rigors’ we all get a bit twitchy ourselves. But is there any evidence to suggest that rigors = serious bacterial infection?
Jed is a thriving 8 month old brought in by his parents with a fever of 38.8oC; he’s tachycardic and grumpy. His fever began twelve hours ago, and aside from crankiness, he is otherwise well. He is fully immunised and there are no sick contacts at home. You get a call from your Mixed Emergency Department intern (it’s their first week) saying they can’t find a source for a fever…
A thorough physical examination is essential to define a source of fever.
Learn and consider the NICE Guidelines traffic light system.
Have a very high index of suspicion in any child under 3 months presenting with a fever.
Common causes of occult infection are bacteraemia, urinary tract infection and pneumonia.
Always think about meningitis & sepsis.
A fever without source is “a fever occurring for less than a week without an adequate explanation after a careful history and thorough physical examination”.
So, for that thorough physical examination, where are the essential places to look? Where should you look for the source of fever?
The infectious sources of fever arise from cavities that you can extract a bodily fluid from. In short, anywhere there is goo! Think of those places where you can get a sample of fluid, and look there.
Firstly, the brain. Think “Does this patient have meningitis?”; it’s a good question to consider this in any paediatric patient. Recall the signs and symptoms and actively seek them out. Examine for photophobia, Kernig’s, nuchal rigidity and lethargy remembering there’s a lower likelihood of these symptoms being present in younger children. Also remember that infants may present as irritable or feeding poorly. Palpate the fontanelle (with the child at 45o) for bulging. Consider the age of the patient; babies under three months are at a much higher risk of meningitis.
Next, think does this child appear septic? Bacteraemia is present in between 2 & 15 percent of febrile children under 3 months. Despite the introduction of the effective 7-valent (and now 13-valent) streptococcus pneumoniae vaccine, bacteraemia still occurs and can lead to focal infections – including meningitis – in around 10% of children, depending on the bacteria. Mortality differs depending on the pathogen, around 1% in pneumococcal, 4% of Neisseria meningitidis; an unidentified bacteraemia can have catastrophic outcomes. Blood cultures are really important if you’re going to start empiric antibiotics!
Urinary tract infections are common in kids! Catching a urine sample can be challenging in a child who’s a bit dry. At the earliest convenience, give the child a wash down and have a parent or keen helper catch a clean urine sample. It’s often unhelpful sending a bag-caught urine for culture, as the false positive rate isn’t worth the hassle (or distraction). Instead, either go straight for a clean catch, in-out catheter or if the child is unwell do a suprapubic aspirate. If the bag urine is already caught, a dipstick for leuks/nitrites may be of use, but ultimately, an uncontaminated sample is essential. Check out the NICE UTI guidelines.
Examine the lungs and have a good hunt for any work of breathing. The constellation of fever, tachypnoeic & cough increases the likelihood of a pneumonic focus. Any changes on auscultation can be suggestive of bronchiolitis, pneumonitis, reactive airways disease, pneumonia or other lung pathology as it relates to the child’s age, so really take your time to listen to the front and back of their chest.
Always examine the ears, nose & throat; it’s a paediatric skill and we see more of these than everyone except the ENTs. Whilst you’re there have a look in the mouth and at the eyes, too!
Nose & rhinorrhoea – copius rhinorrhoea can be a strong sign of upper respiratory viruses. Although a very common cause for fevers in children, beware of ascribing all fevers in a child with rhinorrhoea to a “viral URTI”, without considering the other sources of fever.
Ears – otitis media can cause fevers, or be secondary to a respiratory tract infection. It goes without saying that every child’s ears must be examined. A good pointer is that child with a non-tender pinna and non-tender tragus is less likely to have an otitis externa, but you do need to look.
Throat – Consider the many causes of tonsillitis including Group A haemolytic strep infection. Fevers and a barking cough sounds like croup, but epiglottitis, bacterial tracehitis and inhaled or lodged foreign body can all give you fevers and a toxic appearance. Rarer diagnoses such as a retropharyngeal abscess are worth considering too.
Whilst we’re in the neighbourhood, have a look in the rest of the oral cavity for dental abscesses or lesions. Remember, teething doesn’t give you a fever!
Palpate the lymph nodes – lymphadenitis or malignancy may be hovering in a big, palpable node in the cervical chain, or the inguinal or axillary nodes. The nodes may herald the location of infection, so consider the area of drainage to each node.
In the abdomen – appendicitis, intussusception or gut perforation can make you febrile, as can a raft of other abdominal causes. You must lay hands on the child’s abdomen! Check for hernias, torted testis and palpate the flanks for tenderness. If they have diarrhoea or even just loose stools, catch one – you can send it for rota, adeno and norovirus or other infectious causes of diarrhoea; if there’s blood or mucous, consider haemolytic uraemia syndrome, salmonella, campylobacter and so on.
Thoroughly inspect every inch of the child’s skin – search for signs of local and systemic infection such as pus, cellulitis, rashes, petechiae. Localised infection, such as furunculitis or boils, contaminated lacerations, grazes or puncture sites may be a source for fevers, or entry points for osteomyelitis. Don’t forget to check the back for abscesses!
Document any rashes comprehensively, avoiding diffuse terms like ‘maculo-paupular’ or ‘viral rash’. Say where and what it is; the dermatologists have a superb array of adjectives for use.
Impetigo, varicella zoster or herpes zoster can also cause fevers and systemic illness. Don’t forget the genitals or the perianal area, Group A Strep or foreign bodies can be hiding away there.
Whilst the top and pants are off examine the bones and joints, really noting any red, painful or under mobilised joints. Children are at much higher risk of septic arthritis and osteomyelitis.
In high risk populations, think about rheumatic heart disease and double check those heart sounds.
Eyes – you’ve likely already checked for photophobia early on by flicking on the lights and checking the pupils. If the eyes are mucky, swab them and send it to the lab.
Check the conjunctivae for injection as well as the surrounding tissue for any hint of orbital or periorbital cellulitis. Insect bites or grazes around the area can be the only clue to an entry point.
Remember, in a well-appearing but febrile child, the causes of occult infection to consider are pneumonia, UTI and bacteraemia. Although most children will have a self-limiting, non-specific viral illness, it’s safe and appropriate to consider and exclude bacterial sources. In summary, if there’s goo, look there for the source of that fever. If you’ve looked in all of these places and still have nothing to hang your hat on, then you’re looking at a true “fever without source”. Treat empirically, considering the age and appearance of the child.
This is not an exhaustive list of the causes of fever, it’s more an approach to make sure you’re looking in all the places that fever can be hiding.
Finally, make sure you check out Tessa Davis’ Life in the Fast Lane post on the NICE Guidelines for Fever and when to be really worried.
Tessa Davis via Life In the Fast Lane : Nice Fever Guidelines in Kids
NICE Guidelines – Feverish illness in children
NICE Guidelines – Urinary tract infection in children
A 6-month-old child presents with cough and runny nose for 2 days with increased work of breathing. She is working hard with moderate subcostal recession. Once you hear the cough, it can only mean one thing….the start of bronchiolitis season.
Bronchiolitis is, as the name suggests, inflammation of the small bronchi and bronchioles. The clinical entity we know as bronchiolitis is the most common admission diagnosis in patients under 2, accounting for high morbidity in this population. More succinctly;
“A seasonal viral illness characterized by fever, nasal discharge, and dry, wheezy cough. On examination, there are fine inspiratory crackles and/or high-pitched expiratory wheeze.” – The University of Nottingham, in their 2009 Acute Breathing Difficulty Guideline.
It is most often caused by respiratory syncytial virus (50-80%), as well as parainfluenza (especially PIV3), human metapneumovirus, influenza, rhinoviruses, and adenovirus. In the words of my first general paediatric consultant:
“Bronchiolitis is a typical winter illness, which almost every child gets in their first two years of life. What matters is first, how old you are when you get it, and secondly, what kind of condition you’re in.”
Several days of upper respiratory tract symptoms, including fever, rhinorrhoea, and coryza. These develop into a wheeze, tachypnea, increased work of breathing, moist cough, and fevers. The increased work of breathing often leads to decreased oral intake, with or without dehydration. In infants, poor feeding and apneas with or without cyanosis may be present.
Auscultation of the lungs may reveal a wheeze and transmitted upper airway sounds. A focal zone with decreased air entry or coarse crackles is more consistent with pneumonia.
It is well described that a typical course of bronchiolitis lasts 7-10 days, with night 2-3 being the most severe. In this respect, it is important to reconsider the diagnosis in any patient not ‘sticking to the script’, or deteriorating after initial improvement. A cough may last for up to four weeks.
The diagnosis is primarily clinical. Clinical features of an area of the lung with decreased air entry or consistent focal crackles warrant a chest radiograph to exclude pneumonia. A chest radiograph infrequently adds to the clinical picture and is not routine.
Some clinicians will obtain a nasopharyngeal aspirate or flocculated swab for respiratory virus PCR. There is also a rapid antigen test, which is good for ruling in, rather than ruling out, a particular virus.
Although typing of the respiratory virus causing bronchiolitis may be helpful for bed management and nursing, it is unlikely to alter management.
This point remains controversial as knowledge of the causative organism may allow some prognostication regarding illness course and prevent unnecessary antibiotic usage. It does not rule out dual respiratory tract pathology, such as secondary bacterial infection.
In particular, there is merit to obtaining swabs in patients with risk factors for severe infection, thus:
Around 50% of children with severe infections have none of the above risk factors.
Children at risk of severe infection and mild symptoms should be admitted and observed.
As noted earlier, the course of the illness is that night 2-3 is usually the most severe, thus any child presenting earlier will potentially worsen, sometimes quite rapidly.
Management is supportive, with rehydration and fluid maintenance whilst unable to feed and respiratory support as required. If the child is unlikely to deteriorate and does not require inpatient observation or additional support, they can often be managed in the outpatient setting. It’s essential to provide clear safety-net advice for when to return to ED as well as strongly encouraging an early GP review.
Infants <6/52 and patients with the risks for severe infection above are at risk of apnoeas, so cardiorespiratory monitoring is indicated. Oxygen saturation monitoring is indicated in all children. Keep a close eye on work of breathing and hydration as these can change as the illness progresses.
There are differing opinions regarding target oxygen saturation in patients with bronchiolitis. Most would agree on a target somewhere between ≥94% and ≥92%. But which?
Consider the oxygen-haemoglobin dissociation curve:
Conditions that push the curve to the right (higher PO2 to maintain SaO2 and curve has a steeper gradient), are decreasing pH, increasing temperature.
Thus, in a febrile, acidotic child the curve is pushed to the right, requiring a higher PO2 to maintain the same level of saturation and thus increasing the likelihood of desaturations. In these children, aiming for an SaO2 of 94% would be reasonable. Local guidelines may have differing suggestions.
Oxygen delivery can be via nasal cannulae, or if bronchiolitis severe by high-flow humidified oxygen via nasal prongs. High-flow provides continuous positive airway pressure (CPAP) and aims to avoid intubation.
Flow rates of 2l/kg/min provide a PEEP of 4-8cm, improving the functional residual capacity.
In life-threatening cases, sedation, intubation and ventilation may be necessary and should be done with PICU and senior support. In regional and rural settings, it’s important to be aware of local limitations and consider early transfer to a tertiary centre for deteriorating, high-risk patients. For the care of such patients, make early contact with your regional Paediatric Emergency Transfer Services.
Children with acute illness are susceptible to SIADH and hyponatremia. Thus, it’s important that whilst supporting the increased insensible losses and decreased intake as a result of respiratory distress we don’t over hydrate the patient. Around 2/3 maintenance for an infant not tolerating oral intake is sufficient to both hydrate and reduce the likelihood of SIADH. Most paediatric departments would use 0.9% saline and 5% dextrose.
In mild cases of bronchiolitis, small, frequent feeds may provide sufficient hydration. As respiratory distress increases, nasogastric feeds (continuous vs bolus) may be required. For a child with severe bronchiolitis, on high-flow nasal prongs or moderately dehydrated, intravenous therapy is indicated.
Antibiotics are not indicated for bronchiolitis, nor is there a benefit in the use of steroids, nebulized adrenaline, or bronchodilators. Additionally, there is no benefit in physiotherapy or macrolide antibiotics.
Although salbutamol does not alter the clinical course of bronchiolitis, in the case of a strong family history of atopic disease or asthma it would not be unreasonable to trial a course of salbutamol, particularly in the child >6 yo who is presenting with a wheeze.
It’s important to assess the child pre- and post-salbutamol for both subjective and objective clinical response. This may also have some utility in differentiating between viral bronchitis and a viral-induced wheeze. Again, this is a controversial sub-topic with evolving evidence.
Bronchiolitis has a broad range of differential diagnoses; it is important to consider these, particularly if the patient is not responding as expected to your initial management, or if there are features of the history or examination of an alternative diagnosis.
Pneumonia or other pulmonary infections, including mycoplasma and pertussis – listen for focal crackles or signs, productive cough or radiological features of pneumonia. These patients will usually have a fever, in addition to cough and tachypnoea. Consider viral, bacterial, chlamydial or mycoplasma pneumonia. This is a common alternative diagnosis for bronchiolitis.
Recurrent viral-triggered wheezing – another common alternative diagnosis, viral-triggered wheeze may present quite similarly. Features in the history that might point you towards a viral-induced wheeze, is that of an antecedent URTI which had resolved, followed by wheeze and work of breathing a few days later, with or without fever. Differentiating between RSV bronchiolitis and a viral-induced wheeze can be particularly challenging.
Meningitis – most consultants can describe a patient who presents during a “bronchiolitis epidemic” who has meningitis. Always consider this differential diagnosis. Although it’s had to find case reports in the medical literature, unfortunately, the lay press has many examples of this.
Foreign body aspiration – characterized by rapid onset and failure of initial management. May have a low-grade fever. Have a high index of suspicion.
Croup – usually stridor rather than a wheeze, and with a similar gamut of causative organisms. Laryngotracheobronchitis can present with a wheeze in a child with bronchitic disease.
Aspiration pneumonia – consider more strongly in children with poor airway protection, including spastic cerebral palsy, or in any child having seizures or convulsions.
Gastro-oesophageal reflux – up to 40-50% of infants with GORD present with wheeze or respiratory symptoms. There is also an association with chronic cough.
Asthma – an acute exacerbation is less likely to have fevers, and will often have a personal history of atopy or allergy, or a family history of asthma. It would be rare to diagnose a child under two with asthma at a first presentation to the hospital with a wheeze.
Chronic pulmonary disease – chronic neonatal lung disease or prematurity predisposes the child to respiratory infections.
A mediastinal mass – there are multiple case reports of mediastinal masses presenting with a wheeze and respiratory distress.
Tracheoesophageal fistula – late-presentation of H-type tracheoesophageal fistula may present as coughing, abdominal distension, and recurrent chest infections. Sundar’s well described 1975 case series can be found here.
Congenital heart disease and heart failure – to be considered in any neonate presenting with increased work of breathing, with or without apneas. These patients will look unwell, with a constellation of symptoms including disproportionate tachycardia, poor perfusion with or without cyanosis, weak femoral pulses & murmurs.
Vascular ring, congenital lobar emphysema or a bronchogenic cyst – may also present in neonates with a “bronchiolitis-sounding” history.
Grimes, A. All That Wheezes… HOSPITAL PEDIATRICS Vol. 2 No. 1 January 1, 2012 pp. 47 -50
Marlais M, Evans J, Abrahamson E. Arch Dis Child 2011;96:648-652 doi:10.1136/adc.2010.201079 Clinical predictors of admission in infants with acute bronchiolitis.
Paediatric Grand Rounds – Acute Viral Bronchiolitis in Children by Dr Nitin Kapur, Respiratory Paediatrican. Lecture on 5th June 2013 @ Royal Children’s Hospital, Herston QLD. Slides.
Fitzgerald, D.A. & Kilham, H.A., Bronchiolitis: assessment and evidence based management. MJA 2004; 180(8): 399-404.
Lakhanpaul M, Armon K, Eccleston P, et al. An Evidence Based Guideline for the Management of Children Presenting With Acute Breathing Difficulty. Nottingham, United Kingdom: University of Nottingham; 2002.
Louie, M C & Bradin, S. Foreign Body Ingestion and Aspiration Pediatrics in Review August 2009; 30:295-301.
Sheikh S, Allen E, Shell R, Hruschak J, Iram D, Castile R, McCoy K. Chronic aspiration without gastroesophageal reflux as a cause of chronicrespiratory symptoms in neurologically normal infants. Chest. 2001 Oct;120(4):1190-5. PMID: 11591559.
Sheikh S, Stephen T, Howell L, Eid N. Gastroesophageal reflux in infants with wheezing. Pediatr Pulmonol. 1999 Sep;28(3):181-6. PMID: 10495334.
National Asthma Council Australia. Asthma Management Handbook 2006. Melbourne, 2006.
Phelan, E., Ryan, S. & Rowley, H. Vascular rings and slings: interesting vascular anomalies. The Journal of Laryngology & Otology (2011), 125, 1158–1163.
|Age||if meningitis excluded||if meningitis is NOT excluded||if Staphylococcus aureus suspected|
|< 3 months||Amoxicillin 50 mg/kg q6hrGentamycin 7.5 mg/kg q24hr||Amoxicillin 50 mg/kg q6hrGentamycin 7.5 mg/kg q24hr
Cefotaxime 50 mg/kg q6hr
Consider especially in infants aciclovir 20mg/kg q8hr
|Amoxicillin 50 mg/kg q6hrCefotaxime 50 mg/kg q6hr
Vancomycin 15 mg/kg q6hr
Clindamycin 15 mg/kg q8hr
|> 3 months||Cefotaxime 50 mg/kg q6hrFlucloxacillin 50 mg/kg q4-6hr||Cefotaxime 50 mg/kg q6hrFlucloxacillin 50 mg/kg q4-6hr||Cefotaxime 50 mg/kg q6hrVancomycin 15 mg/kg q6hr
Clindamycin 15 mg/kg q8hr
|any age group, if immuno-compromised||Meropenem 20 mg/kg IV q8hrVancomycin 15 mg/kg q6hr
Gentamycin 7.5 mg/kg q24hr
|if Meningitis is NOT excluded, and suspicion of severe Meningitis (gram stain), replace Flucloxacillin by Vancomycin 15 mg/kg q6hr to cover for Penicillin resistant Pneumococcus Meningitis|
Definition: accumulation of lymphatic fluid or chyle in the pleural space (triglyceride level >1.1 mmol/l or total cell count >1000 cells/ml with >80% lymphocytes).
Definition: oxygen delivery provided by CO does not meet oxygen demand (or CI < 2.1 L/min/m2) due to inflammatory response of CPB, myocardial ischaemia from aortic crossclamp, hypo-/hyperthermia, reperfusion injury and surgical treatment (ventriculotomy).
You’re working as the paeds reg overnight at a regional centre when ED phones you about an incoming patient – Josef, 8 – who’s in a delayed intensification cycle of his treatment for ALL and has a fever of 38.6oC. Josef’s last chemotherapy was last week, and an FBC done 2 days ago showed WCC 4.0×109/L with an absolute neutrophil count 0.9×109/L.
Febrile neutropaenia may be the only feature of a life-threatening infection, a major cause of morbidity and mortality in paediatric oncology patients. A 2005 study of over 12000 children established a mortality rate as high as 3%. Around 1 in 5 children will have microbiologic evidence of infection during induction chemotherapy, and this number jumps to ~40% if the fever returns upon ceasing antibiotic therapy.
You move Josef to a resus bay with isolation (and cytotoxic) procedures in place. He is febrile to 38.6oC, mildly tachycardic but normotensive. You do not identify an immediate threat to life after considering shock, overwhelming sepsis, respiratory compromise. You take a thorough history and examine Josef.
Neutropenia is an absolute neutrophil count <0.5×109/L, or <1×109/L with a predicted decline to less than 0.5×109/L within 48 hours.
What is the child’s oncology diagnosis and where in the treatment course are they? Different chemotherapeutic agents have relatively stronger myeloablative effects, leading to more fulminant and predictable neutropenia. As chemo agents vary with diagnosis and cycle, it’s important to clarify which medications have been given and the number of many days since last chemotherapy.
Compliant with antifungal & pneumocystis prophylaxis? Think about PCP pneumonia in any oncology child presenting with work of breathing.
What kind of central venous access +/- last accessed? Central venous access is a double-edged sword – an essential access for chemo that allows a reduction in the number of peripheral venipuncture, but also the most common source of bacterial infection in chemo kids.
Sick contacts? Remember, just because your patient has an oncology diagnosis doesn’t mean they don’t catch other age-appropriate illnesses, like gastroenteritis, upper respiratory infections and the like. Of course, they’re often more severe, but it’s important to look!
Examine for signs of dehydration, sepsis and anaemia.
Examine the central line site. A good time to look is when the line is being accessed for cultures. Check the age of the dressing and note any erythema or cracks in the line.
Look at the skin all over.
Examine as for a fever without source.
Have a good look in the mouth – mucositis is common and a possible entry site.
Likewise, the perianal area is susceptible to skin breakdown, with or without perianal abscesses.
Take particular note of any areas of erythema.
It’s worth considering the pathophysiology of erythema; local inflammatory mediators (IL-1, IL-6, TNFa) signal neutrophils to marginate, roll and undergo diapedesis to the area of action. But in the absence of a full neutrophil response, any localized erythema will likely be reduced, and pus not formed in the usual volume.
Thus, the smallest area of erythema should be considered as a possible source for infection, especially around central access sites or surgical wounds.
FBC (are they actually neutropaenic?)
Blood culture (central lines/PICC)
Urea & electrolytes, consider calcium, magnesium, phosphate (dehydration, renal dysfunction, tumor lysis syndrome)
Liver function tests (liver dysfunction secondary to chemotherapy agents)
CXR if increased work of breathing, poor SpO2
Nasopharyngeal aspirate if rhinorrhoea (make sure the PLT >50×109/L beforehand)
Consider coagulation profile
Focused investigations as per history and examination
Stool sample if diarrhoea, with C. difficile toxin if on recent antibiotics
IV access is gained, via Josef’s tunneled central line, by an experienced staff member. Bloods are sent for FBC, culture and UEG, LFT, Ca/Mg/PO4, coags.
Most hospitals have well-established protocols for the treatment of febrile neutropenia. Be aware of where to find yours and the choice of anti-infective agents.
Start antibiotic treatment promptly; it may be life-saving. This is not a time to faff about waiting for the results of investigations as antibiotics are the treatment irrespective of any preliminary results.
Some protocols advise anti-fungal treatment in addition to antibiotics. These protocols will vary between centres and over time with changing resistance patterns.
Remember to discuss your patient with the oncologist on call; these kids will usually need admission and, on occasion, transfer to the tertiary oncology centre.
Antibiotics are the mainstay of treatment in febrile neutropaenia. Miadema and her Dutch colleagues are presently undertaking a Cochrane Review of intravenous vs oral empiric treatment of febrile neutropaenia.
The Therapeutic Guidelines currently recommends:
Piperacillin+tazobactam 100+12.5mg/kg (Max 4+0.5g) IV q8h
or cefipime 50mg/kg (Max 2g) IV q8h
or ceftazidime 50mg/kg (Max 2g) IV q8h
If you have a suspicion of MRSA, central line infection or haemodynamically unstable, add vancomycin. Gentamicin or amikacin may be indicated.
Management of Fever in the Paediatric Oncology Patient v3.0 17102012 Febrile Neutropenia Protocol QPHON QCCC
Miadema et al. [Protocol] Empirical antibiotic therapy for febrile neutropenia in pediatric cancer patients. Cochrane Library.
Afzal, S. et al. Risk Factors for Infection-Related Outcomes During Induction Therapy for Childhood Acute Lymphoblastic Leukemia, The Pediatric Infectious Disease Journal • Volume 28, Number 12, December 2009 pp 1064-68
Therapeutic Guidelines : Antibiotic. Severe Sepsis: empirical therapy (no obvious source of infection): febrile neutropenic patients. Therapeutic Guidelines Group. Revised June 2010. (etg40 July 2013)