Searching for sepsis

Cite this article as:
Anna Peters. Searching for sepsis, Don't Forget the Bubbles, 2020. Available at:
https://doi.org/10.31440/DFTB.31160

The child with “fever” is one of the most common paediatric presentations to the emergency department. Most of these children are managed conservatively with parental reassurance and discharged home with a safety net identifying red flags. However, failing to identify those with “sepsis” has devastating consequences. How often do we get it wrong or worry about getting it wrong? We’d all love an evidence-based clear cut path for flagging and managing febrile children at risk of sepsis. Currently the approach in the UK is predicated on the NICE SEPSIS (NG 51) screening system which has anecdotally performed poorly with concerns it is poorly specific (i.e lots of false positives). Nijman and colleagues aimed to objectively assess the impact of the NICE Sepsis screening approach in children.

Nijman RG, Jorgensen R, Levin M, Herberg J and Maconochie IK. Management of Children With Fever at Risk for Paediatric Sepsis: A Prospective Study in Paediatric Emergency Care. Frontiers in Pediatric Care 2020; 8:548154. doi: 10.3389/fped.2020.548154

The lead authors looked at the various warning signs of serious infections in febrile children presenting to PED. Their aim was to then determine these children’s risk of having sepsis and to evaluate their subsequent management.

Who did they study?

Over 5000 children (5156 to be exact) aged 1 month to 16 years old presenting with fever over a period of 9 months from June 2014–March 2015 in a single PED at St Mary’s Hospital, UK were analysed.  Febrile children with no warning signs of sepsis were then excluded from the final cohort. The second largest group excluded from the final cohort was children with a complex medical history (n=119).  The decision to exclude this particular cohort is important given that ‘complex medical patients’ are more likely to have sepsis. The authors make the valid point that this group has features very different from the intended cohort, such as having different management plans in the context of fever. After these exclusions, plus a few further exclusions (lack of consent, lack of complete data or excluded because the child didn’t have any warning signs) the final cohort was of 1551 children. 

What did they do?

They first looked at the numbers of febrile children with tachycardia and tachypnea by using APLS and NICE (the National Institute of Healthcare Excellence) thresholds.  Subsequently, they looked at the numbers of febrile children fulfilling sepsis criteria by using well-known sepsis screening tools (NICE traffic light guidelines, SIRS, qSOFA, Sepsis Trust UK trigger criteria).

All the data for this study (vital signs, clinical signs and symptoms, tests, working diagnosis, need for hospital admission, timeliness of interventions) were collected electronically, having been recorded prospectively for all febrile children.

What did they look for? 

As a primary outcome the study determined:

  1. The incidence of febrile children who present with warning signs of sepsis 
  2. How often these children fulfilled paediatric sepsis criteria 
  3. How frequent invasive bacterial infections (IBIs) occurred in this population 
  4. How frequent PICU admissions occurred in this population.

Secondary outcomes included the compliance of clinicians with the paediatric sepsis 6 care bundle (PS6), what clinical interventions were and were not used from this care bundle and the timeliness of the interventions that were undertaken

What did they find? 

Almost a third of children aged 1 month to 16 years who presented to the PED had fever (28% to be exact).

41% of these febrile children had one or more warning signs (our study population).

The incidence of IBI was 0.39%. Of these children, only 0.3% required PICU admission.

This meant that using the sepsis guideline recommendations, 256 children would need to be treated to catch one IBI. Another way of saying this is the number needed to treat was 256. NNT for any serious outcome was 141.

How did the sepsis guidelines fare?

The thresholds for tachycardia and tachypnoea yielded a high false positive rate.

Adding sepsis criteria to predict the presence of a serious bacterial infection (SBI), IBI or PICU admission was also unreliable, with a lot of false positives.

Lactate levels were not significantly associated with the decision to give IV fluid bolus or presence of SBI, IBI or PICU admission. There WAS, however, a significant association between lactate levels and hospital admission.

Looking at the Paediatric Sepsis 6 Interventions, although many children triggered, two-thirds (65%) of the children with PS6 warning signs had none of PS6 interventions. And when it came to the ‘golden hour? Only a third (36%) of children with IBI or PICU admission received all PS6 interventions in the ‘golden hour with only 39 children (2%) receiving a fluid bolus

What does this all mean?

It is important to note that this study was only conducted in one single PED and in a time period that was before the NICE sepsis guidelines were formally implemented into practice.  The data was collected for this study via an electronic interface. While large amounts of data can be collected rapidly there can sometimes be gaps, either due to extraction issues or brevity on the behalf of clinicians that don’t give a comprehensive picture. Data were also only taken from initial triage and not from any clinical deterioration in the ED.  Given that acuity changes over time, especially in children with fever, this may have missed subsequent clinical change although is a pragmatic approach given the way that sepsis screening tools are applied in nearly all Emergency Departments. 

Numbers needed to treat were exceptionally high. Despite the allure of a protocol-based screening and management pathway,  the benefits of catching true sepsis early must be weighed against the possible unwanted effects of overtreating or overdiagnosing mostly well children in a potentially resource-stretched PED. The study really does highlight the difficulties we face when screening for a septic child in a generally well cohort, the ‘needle in a haystack’.

Essentially, what this study shows us is that serious infections are rare and most children who are categorised as ‘at risk of sepsis’ can in fact be managed conservatively with little intervention other than observation. It is clear that our current guidelines have very poor specificity; and while they tell us to investigate and treat lots of children, a lot of the time we as clinicians choose to rely on our clinical judgement and essentially ‘do nothing’. Observation and good clear red flagging must not be underestimated.  Instead of continuing to research more and better early predictors of sepsis, such as point of care biomarkers, perhaps we should be looking at this from another angle. The focus of the lens can also be flipped; we also need more research on how it can be safe NOT to do anything too. 

We’ll end with some thoughts from the authors

The Infections in Children in the Emergency Department (ICED) study is a single centre, prospective observational study. The study describes unique and carefully curated clinical data of febrile children with warning signs of sepsis, from a period prior to the implementation of the NICE sepsis guidelines. 

Our results confirm what many paediatricians dealing with acutely unwell febrile children already suspected: that many febrile children have warning signs of sepsis, but that the large majority have non-life threatening infections. 

Our findings will hopefully contribute to ongoing discussions about the use of sepsis screening tools in paediatric emergency medicine. Our study makes it clear that current tools lead to a high number of false positive cases, and their usefulness in routine clinical care in paediatric emergency medicine should be questioned. Escalation to senior decision makers of all children with warning signs of sepsis should be aspired, but is seldomly feasible in clinical practice and with unproven impact on reducing missed cases and optimising clinical care for the total cohort of febrile children. 

Although all children with serious infections would have been detected by the various sepsis tools, it is now evident that we need better tools to more selectively identify children at the highest risk of sepsis. Future studies should explore the utility of machine learning as well as the potential of combining clinical signs and symptoms with point of care biomarkers.

Ruud Nijman

Pyrexia of Unknown Origin Module

Cite this article as:
Beatrice Zanetti. Pyrexia of Unknown Origin Module, Don't Forget the Bubbles, 2020. Available at:
https://doi.org/10.31440/DFTB.27689
TopicPyrexia of Unknown Origin
AuthorBeatrice Zanetti
DurationUp to 2 hours
Equipment requiredNone
  • Basics (10 mins)
  • Main session: (2 x 15 minute) case discussions covering the key points and evidence
  • Advanced session: (2 x 20 minutes) case discussions covering grey areas, diagnostic dilemmas; advanced management and escalation
  • Quiz (10 mins)
  • Infographic sharing (5 mins): 5 take home learning points

We also recommend printing/sharing a copy of your local guideline.

To prepare for this session, learners could read the below case report article (20 minutes): 

Wood M, Abinun M. and Foster H., Pyrexia of unknown origin. Archives of Disease in Childhood, Education and Practice, 89 ep 63-69 (2004) 

And/or look at these useful resources from the web: 

https://pedemmorsels.com/fever-of-unknown-origin/ (5 minutes)

https://dontforgetthebubbles.com/recurrent-or-periodic-fevers-investigate-or-reassure/ (10 minutes)

https://dontforgetthebubbles.com/tuberculosis/ (5 minutes)

https://dontforgetthebubbles.com/claire-nourse-tuberculosis-at-dftb17/ (20 minutes)

https://gppaedstips.blogspot.com/search/label/Juvenile%20idiopathic%20arthritis  (10 minutes)

https://www.paediatricfoam.com/?s=kawasaki (10 minutes)

The term pyrexia of unknown origin (PUO) is used when a patient has more than 8 days with fever (temperature> 38°C) without a clinical diagnosis after exhaustive investigations have been carried out (in hospital or in primary care). 

Other more specific PUOs are:

  • nosocomial PUO
  • neutropenic PUO
  • HIV-associated PUO

These 3 have specific risk factors and will not be covered in this session. 

Fever is a sign of an underlying pathology. In PUO, pyrexia is usually the main symptom while other signs may be very subtle. Many times, the underlying disease is a common pathology that is presenting in an atypical or incomplete way.  

Here are 3 main clinical dilemmas for clinicians: 

In the paediatric population, 30% of PUO will not reach a final diagnosis. However, in those cases, PUO is often a self-limited and benign episode. 

When a definitive diagnosis is reached, the majority of causes are related to infectious diseases (38%), followed by connective tissue disorders/autoimmune pathology (13%) and malignancies (6%).

At present, there is not a generic PUO work-up since this wouldn’t be efficient. Remember that more than ¼ of cases are benign and self-limited! 

In many cases, PUO is a consequence of a late diagnosis. Clinical history taking and careful physical examination are crucial to pick up subtle signs and guide the complementary tests and imaging. New signs and symptoms, which weren’t present on initial examination, can appear later on.

In the literature, retrospective studies have shown that when imaging requests are prompted by some examination finding, they are more likely to yield a positive result. 

Here’s a table with possible differential diagnosis based on clinical findings.

Diagnosis based on common clinical findings
Rash
Maculopapular -Purpuric-Erythema Nodosum-Butterfly rashEBV, Kawasaki Disease, SOJIA, Typhoid feverCMV, Endocarditis, Leukaemias, Histiocytosis, VasculitisTB, Ulcerative colitis, Crohn’s disease, Streptococcal infectionSLE, Dermatomyositis
Adenopathies
-Infections-Connective tissue disorders-Malignancies-Other causesCMV, EBV, TB, Bartonella (Cat-scratch disease)Rheumatoid Arthritis (RA)
Lymphomas, Leukaemias, HistiocytosisSarcoidosis and Primary Immunodeficiencies
Splenomegaly
-Infections
-Connective tissue disorders-Malignancies
-Other causes
TB, Bartonella, Malaria, Visceral Leishmaniasis, Endocarditis, Brucelosis, SalmonelosisSLE, RA
Lymphomas, Leukaemias, Histiocytosis, Macrophagic Activation SyndromePrimary immunodeficiencies
Arthritis
-Infections-Connective tissue disorders-Malignancies-Other causesTB, Lyme disease, Brucella, Staphylococcal InfectionSOJIA, RA, SLE, Rheumatic fever
Leukaemias

The speed of the complementary tests will depend on the general appearance of the patient. Empirical treatment with antibiotics can blur the microbiology results resulting in a delayed diagnosis. The empirical use of steroids can mask other pathologies and again delay the diagnosis. 

For the above reasons, clinicians should reserve empirical treatment with antibiotics or steroids to the sick patient based on clinical assessment.

Causes of fever of unknown origin
Infectious: Bacterial infections: Localised: Pyelonephritis, Sinusitis, Mastoiditis, Pneumonia/pleural effusion, Osteomyelitis, Endocarditis, Intravenous catheter infectionAbscesses (intracranial, dental, intestinal, hepatic, pelvic…)
Systemic infections: Tuberculosis, Brucellosis, Bartonella (cat-scratch disease), Leptospirosis, Q fever (Coxiella), Lyme disease, Salmonellosis (Typhoid fever), Tularaemia. 
Viruses: EBV, CMV, Adenovirus, Enterovirus, HIV, Dengue
Fungal: Blastomicosis, Histoplasmosis, Coccidiomicosis
Parasites/protozoos: Malaria, Visceral Leishmaniasis, Toxoplasmosis, Visceral Larva Migrans, Tripanosomiasis
Connective Tissue DisorderKawasaki Systemic Onset of Juvenile Idiopathic Arthritis (SOJIA)Systemic Lupus ErythematousAcute Rheumatic fever
MalignanciesLeukaemia LymphomasSolid tumours (Neuroblastoma)Hemophagocytic SyndromeMyelodysplastic syndromeHistiocytosis 
Other: Drug related feversFabricated illness Inflammatory bowel diseaseCentral origin feverPeriodic feversMetabolic fevers (hyperthyroidism, dehydration)Primary immunodeficiencies

A 14-month-old girl was referred to hospital by GP due to 8 days of fever, non-tender cervical lymphadenopathies (scattered small submandibular, posterior and 1 supraclavicular lymphadenopathies) and mild cough. On examination, the patient has a good general appearance with a mildly red throat and the above described lymphadenopathies. Father is concerned as the child also had a febrile illness the previous week which was labelled as a viral infection. 

Blood tests showed raised WCC (24×109/L) with neutrophilia (18×109/L). Normal lymphocytes (6 x109/L) with a CRP of 30 mg/L. Chest x-ray showed a bilateral bronchial opacification.

The patient was admitted and started on amoxicillin and azithromycin PO. 

Despite 5 days of treatment, the patient is still spiking fevers (see chart below). Blood culture is negative. Clinically stable, cough has now disappeared. You are classifying this patient as PUO. 


What questions do you want to ask the parents? Take a detailed history. 

Why is this patient not getting better despite treatment? 

What investigations can be prompted by clinical findings?

At this point, would you escalate the antibiotic treatment?

In PUO, a detailed clinical history is the most important diagnostic tool that can guide all investigations. Instead of ordering random tests, ask more questions!

When taking the history, consider: 

  • Characteristics of the fever: when did it start, duration and intensity. Note that this child had a previous febrile illness which can be part of the same illness. 
  • Pattern of fever: there are several patterns of fever which can help with the diagnosis. If managed in an outpatient setting, ask the family to do a symptoms diary. From looking at the fever chart, the child has an intermittent fever pattern.
Type of feverCharacteristics of feverPossible causative agent
Intermittent feverSharp febrile peak which goes back to baselineBacterial infections, TB, SOJIA
Remittent feverFever peaks and elevated basal temperatureViral illness, Endocarditis, lymphomas
Sustained feverPersistent fever with minimal variationTyphoid fever, Brucellosis
Recurrent feverPeriods of fevers intercalated with afebrile periodsMalaria, Lymphomas, Borellia
Periodic feverFebrile periods are intercalated with afebrile periods with a predictable pattern of 6 monthsPeriodic fever syndromes 
  • Age of the child: PUO in young children is often caused by infections while in older children and teenagers tends to be caused by a connective tissue disorder or a malignancy.
  • Associated symptoms and signs: headaches, vomiting and diarrhoea, rash, arthralgias, myalgias, bony pain, lymphadenopathies. They can be very subtle therefore a systematic review of all systems is necessary. 
  • Systemic symptoms: fatigue, anorexia, weight loss, sweating.
  • Previous medical history: a history of many bacterial infections can be related to a primary immunodeficiency. The most common primary immunodeficiencies are: 
  1. Common variable immunodeficiency
  2. Chronic granulomatous disease

Usually the immunodeficiencies are associated with complicated infections, failure to thrive, atopic disease or autoimmune disease.

For more information: https://dontforgetthebubbles.com/ent-infections-immunodeficiency/

  • Vaccination history: patient is fully vaccinated as per UK protocol. This includes BCG due to risk factors (Parents are from a country where incidence of TB is > 40/100,000 or more). 
  • Regular medications or any exposure to new medication (think about drug- related fevers).
  • Family History: Ethnic background, consanguinity. Family is from India and there is no consanguinity. 
  • Environmental risk factors: TB contacts, area where family lives, exposure to animals, vectors (mosquitos, ticks…), food intake (unpasteurised dairy products, uncooked meat and fish), international travels (place, malaria prophylaxis and compliance of prophylaxis). 

Patient travelled to India to visit grandparents when she was 9 months old. She was exposed to some mosquito bites.  She lived in an urban area for 3 months. Parents were not aware of any TB contacts.

For international travellers, the following website provides relevant information on potential risks and outbreaks occurring on each country: https://travelhealthpro.org.uk/

When a patient has received a provisional diagnosis plus empirical treatment for at least 48 hours and there is no improvement, clinicians should use a systematic approach to understand the reasons behind the poor response to treatment. 

Here are the four main questions to be answered: 

  1. Is there a problem with the medication? 

The diagnosis is right but the problem is within the treatment. Issues with the treatment could be related to: 

-Drug resistances (MRSA, ESBL bacteria).

This website will provide a map with all antibiotic resistances over the world:  https://resistancemap.cddep.org/AntibioticResistance.php, treatments should, when possible, be guided by microbiological culture results.

-Underdosages of the antibiotics which don’t reach the effective concentration.

-Very virulent bacteria creating a toxin that requires more antibiotics (eg. Staph. aureus PVL) 

-Adherence to treatment (low compliance) 

-Malabsorption of medication (for example vomiting, diarrhoea when taking oral medications)

-The selected antibiotics are not reaching the right place of infection (bone, abscess…)

2. Are we targeting the wrong bug?

Antibiotics are mainly covering for bacteria but the actual infection can be caused by other microbes like viruses, atypical bacteria, TB, parasites and fungal infections. 

3. Is there a problem with the host? 

Consider whether the episode could be only a prolonged febrile syndrome for a common disease due to host problem.  The problem can reside in the immunity (immunodeficiencies), structural problems that can predispose to localized infections (for example a patient with previous abdominal surgery who now has an abscess) or whether the patient has a foreign body or a central catheter that can be the source of the infection.

4. The problem is not infectious: fever can be a sign of a tissue connective disorder, malignancies and other illnesses like a central origin related fever, drug related fever or a factitious illness.

This child has now had at least 12 days of fevers. It could be even longer if we consider that he had a febrile illness labelled as “viral” before this episode. 

On examination, the main clinical sign are the small cervical lymphadenopathies bilaterally with 1 small supraclavicular lymphadenopathy. This clinical finding could prompt the clinician to investigate for others cause: URTI, EBV, CMV, pneumonia, pleural effusion, TB, Bartonella, connective tissue disorders, malignancies, histiocytosis.

After 18 days of intermittent fever, cervical lymphadenopathies and some fatigue, the patient underwent a fine-needle aspiration of the supraclavicular lymphadenopathy. The histology showed a caseating granuloma and the microbiology sample showed acid fast bacilli. TB GeneXpert of the sample and culture were positive for non-resistant Mycobacterium Tuberculosis. Patient was diagnosed with tuberculous cervical lymphadenitis (extrapulmonary TB).

Probably not. Use of antibiotics can delay microbiological diagnosis since the blood cultures’ yield is decreased. If the patient has a good general appearance and fevers are well managed with PRN antipyretics, then clinician can consider withholding the antibiotics until a definitive diagnosis is reached. 

Reaching the definitive diagnosis: 
After 5 days of treatment with amoxicillin and azithromycin and no clinical improvement, basic investigations were repeated and further were added.

Bloods tests
WCC 22 x109/L with neutrophils of 18 x109/L,
Lymphocytes 4 x109/L, CRP 36 mg/L
ESR >60 mm/h
Normal renal and liver function
Blood film normal: No reactive lymphocytes, no lymphoblasts seen
Blood cultures Negative.
CMV Serology: IgM negative, IgG Negative
EBV serology: IgM negative, IgG Negative

Microbiology
Mantoux /TST (Tuberculin skin test): 10 mm induration
(Patient received BCG vaccination)
IGRA (Interferon gamma release assay): Positive
Gastric aspirate for AFB smear: negative 
Gastric aspirate for TB GeneXpert: negative

Imaging
Repeated chest XR: similar features compared to previous one, peri-bronchial shadowing.
Ultrasound of lymph-nodes: several lymph-nodes with nodal matting and surrounding soft tissue oedema. Prominent vascularity in the hilum.

After 18 days of intermittent fever, cervical lymphadenopathies and some fatigue, the patient underwent a fine-needle aspiration of the supraclavicular lymphadenopathy. The histology showed a caseating granuloma and the microbiology sample showed acid fast bacilli. TB GeneXpert of the sample and culture were positive for non-resistant Mycobacterium Tuberculosis. Patient was diagnosed with tuberculous cervical lymphadenitis (extrapulmonary TB).

  • Systemic symptoms 
  • Supraclavicular lymph-node
  • Firm and/or fixed lymph-nodes
  • CXR changes, abnormal Full blood count or increased ESR
  • Adenopathies > 1 cm in a neonate 
  • Suspicion of TB 
  • Persistent lymphadenopathies for more than 4 weeks despite antimicrobial treatment
  • Sometimes in acute lymphadenitis if a patient is not responding after 48 hours of treatment

To note, the patient had received the BCG vaccine. However, it has about 50% efficacy which implies that patients with BCG vaccination can still have tuberculosis. BCG is more effective in preventing children from developing disseminated (Miliary) TB or TB meningitis. She was probably exposed to TB and became infected while in India, subsequently developing the disease over the next few months. 

Contact tracing of family members is mandatory to identify the source case. Usually, children are not very infectious since the majority of cases tend to be paucibacillary (low bacterial load) unless they have lung cavities or extensive lung involvement. 

TB in children often presents in a non-specific way. The typical symptoms are weight loss or failure to gain weight, fever, night sweats and fatigue. When children present with pulmonary TB, this is usually confined within the intrathoracic nodes. Patients may have persistent cough and asymmetrical and persistent wheeze caused by airway compression due to enlarged tuberculous peri-hilar nodes. 

Chest XR can be helpful in the diagnosis of early primary infection by detecting intrathoracic lymph-node enlargement. However, these changes may be subtle as a strong index of suspicion is required. More information on radiological features of paediatric TB can be found on the following link: doi: 10.1101/cshperspect.a017855

Sputum and gastric aspirate mycobacterial cultures have a low diagnostic yield since most children have paucibacillary TB. Recently, diagnostic sensibility for these samples has increased due to the rollout of new molecular techniques (GeneXpert TB PCR).

TST (Mantoux test) and new immunological assays such as IGRAs detect exposure. TST is performed by injecting 0.1ml of tuberculin purified protein derivative (PPD) intradermally into the inner surface of the forearm. The skin reaction produced by the PPD should be read between 48 and 72 hours. The reaction is measured in millimetres of induration, not redness. There are different measures to define a positive result depending on patient background history (for example BCG vaccination) and there are also many causes of false positive and false negative results.For more information (https://www.cdc.gov/tb/publications/factsheets/testing/skintesting.htm).

On the other hand, IGRA is a blood test which measures the body’s immune response (interferon-gamma production) to TB antigens. Our patient had a positive Mantoux test (10 mm) but the result might have been affected by previous BCG vaccination.  However, this result, combined with a positive IGRA, demonstrated that the child had been previously exposed to TB. Unfortunately, neither of these tests can distinguish between latent infection and active disease. 

The patient was treated with Isoniazid (with Pyridoxine), Rifampicin, Ethambutol, and Pyrazinamide for 2 months and Rifampicin and isoniazid for another 4 months. Corticosteroids were not deemed necessary in this case since the lymphadenopathies were not compressing other structures. Empirical treatment of tuberculosis is usually limited to clinical cases where milliary or CNS TB are suspected, as a treatment delay in these cases will often lead to worse outcomes.

3-year-old boy with a 5-day history of fever and loss of appetite presented to the emergency department with his mother as he had been crying all night and refused to put his T-shirt on. No history of trauma reported. On examination, he looked skinny and he was crying when the right arm was moved. Bloods test showed
Hb 9 g/L
WCC 4 x109/L
Neutrophils 1.5 x109/L
Lymphocytes 2.5 x109/L
Platelets 120 x109/L.
CRP 40 mg/L.

Right arm x-ray was normal. The patient was admitted for observation. On the ward, it was noted that he was spiking fevers every night. 

After 3 days of admission, MRI of the right upper limb was performed. MRI showed possible osteomyelitis of the right distal clavicle.  He was diagnosed with acute pyogenic osteomyelitis and was started on ceftriaxone 50mg/kg IV OD. Blood cultures (taken before administration of antibiotics) were negative. Fever settled after 5 days of antibiotics. Patient was discharged home on oral antibiotics for 3 weeks.  

10 days later, the patient was reviewed in the clinic. Mother was worried since the patient had had fevers again over the last 2 days, felt fatigued and was reluctant to walk.

At this stage, what is the differential diagnosis? 

What investigations would you perform? 

What treatment would you give? If you were to suspect an autoinflammatory disease, would you give steroids? 

What is the role of PET-CT in PUO?

Infectious diseases: 

Osteomyelitis: 

Every time a patient presents with reduced range of movement due to a bony pain, osteomyelitis should be considered. In non-verbal children, it can present with irritability and inability to bear weight. It usually affects the metaphysis of the long bones (femur, tibia…). Therefore, the original diagnosis of clavicle osteomyelitis was quite rare. Now that the patient presented again with fever and a new similar problem after receiving adequate antibiotic therapy, another diagnosis should be considered. 

Septic Arthritis: 

It has a similar presentation to osteomyelitis but usually the joint is swollen, red and hot. Ultrasound of the joint can detect joint effusion which can be a sign of septic arthritis. Urgent orthopaedic referral for aspiration +- surgical washout is necessary. 

Connective tissue disorder: 

Transient synovitis: Fever and inability to bear weight can be a common presentation for transient synovitis. In this particular case, the initial diagnosis might have been wrong and the first inflammatory/infectious process could have triggered the production of antibodies causing inflammation over the joint. 

SOJIA (Systemic Onset of Juvenile Idiopathic Arthritis): Usually the joints affected by the arthritis are hot, tender and erythematous. We can suspect this pathology when there are different joints affected at different times. It is usually associated with systemic symptoms (fever and salmon pink rash, splenomegaly, serositis).

Diagnosis is made by elevated ESR, elevated ferritin, absence of antibodies, rheumatoid factor negative and exclusion of malignancy or infectious process.

CRMO (Chronic Recurrent Multifocal Osteomyelitis): this is an idiopathic inflammatory bone disorder with chronic multifocal bone pain. Sometimes systemic symptoms like fever can appear. Clavicle involvement is characteristic of this pathology.

Diagnosis is made by lesion’s biopsy: this will show an inflammatory reaction with no microbiological growth. 

Malignancies: 

Leukaemia: can present with bony pain and fatigue, lethargy and weight loss. Pancytopenia and blast can be seen in the blood film. LDH and uric acid are raised. Definitive diagnosis is reached with the bone marrow aspirate and flow cytometry.

Neuroblastoma: This is a malignancy that usually presents with abdominal mass. Sometimes mass can be found in the thoracic cavity. It appears in children below 5 years. The neuroblasts infiltrate the bone marrow. Therefore, patients can present with bony pain and pancytopenia. 

Diagnosis is reached by abdominal ultrasound and further imaging to evaluate the stage of the disease. Bone marrow aspirate is necessary along urine Vanillylmandelic Acid (VMA). 

Bone tumours (osteosarcoma /Ewing’s Sarcoma): 

Even though bone tumours are much less common than leukaemia and neuroblastoma, the presence of bony pain and prolonged fevers would prompt the diagnosis. LDH is usually elevated with raised calcium. 

X-Ray show bony abnormalities and further imaging with MRI or CT can provide more information. Sometimes biopsy of the lesion is necessary to confirm the diagnosis. 

Bone marrow is recommended in Ewing’s sarcoma. Metastasis and benign bone tumour should also be considered in the differential diagnosis.

Other: Histiocytosis:  this is a systemic illness which can affect bones. Associated symptoms are erythematous skin lesions, oral ulcers, lymphadenopathies, cough, shortness of breath, hepatosplenomegaly, malabsorption. Diagnosis is reached by seeing Langerhans cells in the biopsy of the lesion.

At this point, the patient has had a fever on and off for more than 3 weeks. A provisional diagnosis of osteomyelitis was made based on imaging findings. However, treatment is failing and the patient is now presenting with new symptoms (unable to bear weight). 

Repeated basic investigations:

Full blood count Hb 8.5 g/LWCC 1.0 x109/LLymphocytes 9.0 x109/LPlatelets 100 x109/LBlood cultureNegative
Peripheral blood film NormalUrine cultureNegative
CRP15 mg/LMantoux testNegative
ESR>40 mm/hHIV serologyNegative
Renal functionNormal rangeSickle cell testNegative
Liver function Normal range Chest XR Normal.
LDH900 U/L (240-480)
Uric Acid 12 μmol/L (High)

If after the above investigations, the clinician does not reach a diagnosis, then: 

  • Re-take a good clinical history 
  • Re-assessment of the patient
  • Withhold current medications
  • Do specific imaging (XR/Ultrasound/MRI of the new affected area)
  • Perform immunological studies: rheumatoid factor, ANA and anti-DNA antibodies, Immunoglobulins
  • Perform a bone marrow aspirate and trephine for histology, cytology and microbiology.

In this case, the diagnosis of osteomyelitis was discarded. SOJIA vs Leukaemia were the 2 main differential diagnoses.  Discussion regarding therapeutic steroid treatment for SOJIA was raised. 

Usually in PUO, steroid treatment should be avoided until malignancy is ruled out.  Steroids are used therapeutically in many oncology protocols. The use of steroids in an unconfirmed case of leukaemia can improve symptoms but it can blur the histological picture required for the diagnosis and confuse the staging process. This would lead to a delayed and potentially incorrect treatment. 

It is crucial to perform a bone marrow aspirate before steroid treatment is given, especially if there are symptoms and signs compatible with malignancies. 

In the above case, the full blood count showed mild pancytopenia which can be related to a bone marrow infiltration. The peripheral blood film was normal. Finally, the patient underwent a bone marrow biopsy and this confirmed the diagnosis of T-cell ALL. 

PET-CT is an imaging technique that localises anatomical parts with high metabolic activity, detecting hidden infections, malignancies or any inflammatory foci. 

PET-CT has proven to be useful in patients with PUO who are generally unwell, sick-looking, since early diagnosis is urgent in those patients. Otherwise, PET-CT can be used in those patients who have had extensive investigations done, have not had clinical improvement and still no diagnosis has been reached.

4-year-old boy presented with 5 days of fever, diarrhoea and vomiting and abdominal pain. No relevant past medical history. Fully vaccinated, BCG not included.

Initial blood test showed WCC 24.5 x109/L with neutrophils of 18 x109/L. CRP 139 mg/L. Hb 110 g/L and Platelets of 395 x109/L. He was admitted and started on amoxicillin, gentamicin and metronidazole. Blood cultures were negative and urine culture showed a sterile pyuria (WCC 2250 with no growth). Stool sample was negative. Abdominal ultrasound showed free fluid in the right iliac fossa. On examination, his abdomen was soft with some tenderness in lower quadrants.  He had a second ultrasound which showed findings suggestive of an appendicular mass. A repeated urine sample had 64 WBC and no growth. 

Meanwhile, fevers persisted: on day 7, he was changed to piperacillin-tazobactam and gentamicin. He underwent a laparoscopic appendicectomy on day 8. After operation, he was afebrile for more than 48 hours and antibiotics were stopped. Histological results of the appendix were normal. On day 12 of admission, the patient started again with fever and no focus on examination.

Now that the fever has restarted, and considering the previous history, what investigations would you ask? 

Would you re-start antibiotics? 

Looking at the pattern of fever below, what can you observe? 

Would an echocardiogram help in reaching the final diagnosis?

You should probably start by repeating basic investigations. Results: raised WCC with neutrophilia and thrombocytosis. Hb 101 g/L, WCC 32 x109/L with neutrophils of 24 x109/L, Platelets of 961 x109/L. He had normal renal and liver function.

Infectious diseases investigations: 

Microbiology cultures: Blood cultures were negative, even the prolonged culture for atypical bacteria. Stool sample was negative for viruses, bacteria and parasites. Urine sample became negative (previous sterile pyuria)

Toxoplasma serology:  IgG and IgM negative

CMV serology: IgM positive and IgG positive.  Second sample sent for CMV IgM negative. CMV PCR was negative. The initial positive IgM CMV was considered to be a false positive. IgM positivity in virology/microbiology assays may be non-specific, in patients with autoimmune diseases, cross-reactions.

EBV serology: IgM and IgG negative. 

Blood PCR for EBV, CMV and adenovirus negative.

Respiratory sample PCR: negative.

Lumbar puncture: LP was performed on day 14 of admission: WBC < 3/mm3. RBC <3/mm3. Viral PCR for enterovirus, parechovirus, mumps, VHS1&2 and VVZ negative. Negative culture.

Interferon Gamma Release Assay for TB negative. 

Inflammatory conditions investigations: 
Faecal calprotectin: negative. Since the patient had gastrointestinal symptoms and fever, Inflammatory Bowel Disease (IBD) should be considered as a potential differential diagnosis.

ESR: 50 mm/h.

Ferritin 222 ng/mL: important marker for inflammation. Especially high in Hemophagocytic lymphohistiocytosis /Macrophage Activation Syndrome.

Malignancies: 
Blood film: no atypical cells. Polychromasia and raised platelet count. Neutrophilia.

LDH 446 U/L (high)

Imaging
Day 12 Chest XR: normal
Day 13 Abdominal Ultrasound: normal kidneys and bladder. Normal liver, spleen, gallbladder and bile ducts. No abnormal masses or bowel wall thickening. Trace of fluid in right iliac fossa postoperatively. No fluid collection. 

The patient was clinically stable, so it was decided to wait and hold antibiotic treatment. The patient continued to have daily fevers up to 39°C. On day 14, he had one bilious vomit and became more lethargic therefore antibiotics were restarted (Piperacillin-Tazobactam and gentamicin). The following day, he underwent a Bone marrow aspirate and MRI under general anaesthesia with results as below: 

BMA: Trilineage haematopoiesis. No evidence of abnormal infiltration. No increased haemophagocytic activity. Appearances in keeping with a reactive marrow. Negative for AAFB, both microscopy and culture. 

Abdominal MRI:  There is moderate distention of the proximal small bowel with an apparent jejunal transition point due to ileus, adhesions or oedema from handling.  Some free fluid but no abdominal collections.  No retroperitoneal collection.  No bone marrow abnormality.

On day 16, he did not spike any temperatures.  After 48 hours (on day 18 of admission), he had an evening temperature of 38.5°C.

The patient had 2 episodes when he was apyrexial:

  • the first one between day 9- 11 after antibiotics were escalated (Piperacillin-Tazobactam and Gentamicin (D7)) and after surgery under general anaesthesia (D8).
  • The second afebrile period was on day 16-18 after being re-started on Piperacillin-Tazobactam and Gentamicin (D15) and after he underwent a procedure under general anaesthesia. 

In the first episode, the lack of fever was linked to a good response to antibiotics whereas in the second episode given the fact that a non-infectious condition was highly suspected as a differential diagnosis, the afebrile episode could be linked to the anaesthesia.

A very important investigation to perform in PUO is an echocardiogram to rule out infective endocarditis. In this case, there were no positive cultures or risk factors to point towards an infective endocarditis but it would be useful to rule out this disease. Echocardiography can also help to diagnose Kawasaki disease. In this particular scenario, it would be an incomplete case of KD. 

Reaching the diagnosis: 
On day 19, the patient had an echocardiogram which revealed dilated circumflex artery and an aneurysm of the left anterior descending artery. This finding confirmed the diagnosis of Incomplete Kawasaki. The ophthalmology review showed no pathological findings.  

DIAGNOSTIC CRITERIA FOR KAWASAKI DISEASE
Full case of KawasakiIncomplete case of Kawasaki
Fever (>38°C) every day for 5 days        +At least 4 of the following 5 featuresNon purulent bilateral conjunctivitisCervical lymphadenopathyPolymorphous rashLips/oral mucosa involvementFingers/toes: acute erythema and oedema of palms and soles and then peeling.
Or positive echocardiogram at any time with less than 4 features.
Fevers (>38°C) every day for 3 days+ less than 4 features but diagnosis supported by: Lack of alternative diagnosis (lack to respond to antibiotics, no other pathogen found)High inflammatory markers (high CRP, ESR, NeutrophiliaPresent of other clinical features: Irritability without CNS infectionBCG scar inflammationOther system involvement: CSF pleiocytosis, uveitis, arthritis, gastroenteritis, myocarditis, dysuria, sterile pyuria.

In our particular case: the patient had prolonged fevers with high inflammatory markers (CRP, ESR, Neutrophilia), irritability without CSF infection, sterile pyuria, low albumin, anaemia, thrombocytosis and lack of alternative diagnosis. Furthermore, he had a characteristic echocardiographic finding of Kawasaki Disease.

Patient was started on IVIG and aspirin. Steroids were included in the treatment since the patient already had evolving coronary and or peripheral aneurysm.

For more information on criteria for steroid use in Kawasaki disease, you can read: Eleftheriou D, et al.Managment of Kawasaki disease. Arch Dis Child,99,1 2013 

With regards to the antibiotics, gentamicin was stopped while Piperacillin-tazobactam was continued while evaluating response to IVIG. Piperacillin-tazobactam was stopped after 48 hours. 

Kawasaki disease is rare but early diagnosis is important to avoid cardiological sequelae. Incomplete Kawasaki can present a clinical challenge to diagnose.

You are in an Ethiopian rural hospital. A 7-year-old boy presents to clinic severely malnourished (marasmic type). Mother is complaining of daily fevers for an unknown period of time. 

Patient has cerebral palsy due to an obstructed labour resulting in hypoxic-ischaemic injury. He was in hospital for some time after delivery. He is not vaccinated. He is on phenobarbitone 100mg OD PO for seizures. 

You admit the child to the malnutrition ward and start the appropriate treatment with F-75 Milk. Part of the SAM protocol (Severe Acute Malnutrition) includes a course of at least 7 days with Amoxicillin.  On examination, the patient has a papular rash over hands and groin compatible with scabies but no other clinical findings. On the ward, he spikes a high temperature (39°C) and he is shivering. 

Available investigations at your hospital are performed:

Blood tests: 
Hb 9.1 g/LRenal function and CRP not available in this setting.
WCC 12 x109/L with neutrophils 8 x109/L and lymphocytes 4 x109/LUrine dipstick: leucocytes and nitrates positiveUrine microscopy: many white cells. No culture available.
Platelets 300 x109/LStool: negative for parasites
Blood film: No parasites seen
GGT 61 IU/LHIV antibodies negative
GOT 72 IU/LHepatitis B and C antibodies negative
Bili < 0.5  μmol/L

Based on the above clinical picture and results, what is your differential diagnosis and management? 

Patient was empirically treated but fevers persisted. Given his background of CP and the geographical area, what other infections would you consider?

What other non-infectious causes should be considered? How can you reach the diagnoses in this low-resource-setting?

This is a very challenging patient. Due to their reduced ability to communicate and cognitive impairment, these children are difficult to assess. Furthermore, this patient is malnourished which increases the risk of infections. 

The above results showed a possible UTI which is in keeping with the clinical picture (high fevers, shivering in a patient with high risk of UTI due to his cerebral palsy and poor bladder control). Antibiotics were changed from amoxicillin to amoxicillin-clavulanic to give broader cover for gram negative bacteria (E. Coli, Klebsiella…).

To note, the patient has scabies which is a very common parasitic skin infection that affects mainly the palms and soles and the groin area. If the patient has been scratching over the genital area, it could have triggered a UTI. Furthermore, there are poor hygiene conditions in the area with limited access to water.

The slightly raised GGT and GOT was correlated to the use of phenobarbitone. The hepatitis B and C were negative but the hospital did not have the test for hepatitis A. Nevertheless, the clinical symptoms were not fitting with hepatitis A. 

The patient was treated with co-amoxiclav for 7 days. He initially improved and fevers were spacing in time. However, on day 9 he started again with very high fevers and shivering. He was looking unwell during the fever episodes so he was started on ceftriaxone IV. His baseline temperature was always raised, he had abnormal movements and was irritable. Temperature persisted despite treatment

Another urine sample was requested to rule out a UTI due to a resistant bacterium, since microbiological cultures were not available in the rural hospital. The urine microscopy was negative for WCC and urine dipstick did not show any abnormalities.

Another important differential diagnosis was meningitis. Patient was irritable, had abnormal movements and a fever. The abnormal movements consisted of small twitching of the arms while crying inconsolably. There were considered either shivering or behavioural but there was a lot of discussion if those movements could represent a seizure event. Furthermore, mother was unable to describe the usual seizures that he had at home. The team subsequently realised that there was an error with the regular medications: he was prescribed 100mg of phenobarbitone but mother clarified that at home he was taking 200mg, therefore his daily phenobarbital dose was increased to 200mg OD. To note, the patient did not have any devices (VP shunt) which could increase the risk of infections. In this rural setting, clinicians were not able to perform a lumbar puncture due to lack of laboratory equipment, so the patient was started on ceftriaxone high dose empirically. 

Pneumonia can be a common cause of infection in patients with cerebral palsy since they can have drooling, unsafe swallow prompting for aspiration. Usually, pneumonia in these children can be very silent. In addition, poor nutritional status can increase the risk of severe pneumonia. Patient was not desaturating or with respiratory symptoms but a chest XR was done (in a private clinic) and no lung abnormalities were detected. Furthermore, based on local antimicrobial resistances, the antibiotics he received earlier should have been covered for the most common bacteria causing pneumonia. Gastric aspirate for GeneXpert MTB/Rif was negative. 

Dental infections with abscesses can also present with fever and no other major symptoms. The patient had poor oral hygiene plus the lack of proper tongue movement, drooling and lack of routine dental care made him more prone to these types of infections. These infections are mainly due to anaerobes which should be covered by amoxicillin-clavulanic or ceftriaxone. On examination, no suspicious dental masses were found. 

Viruses can also cause non-specific symptoms. However, they shouldn’t last for very long. He did not have any gastrointestinal symptoms or respiratory symptoms. No palpable lymph-nodes. Unfortunately, in the hospital there were no laboratory diagnoses for viruses.  Full blood count differential was never lymphocytic. 

The most common parasites in this rural area are intestinal parasites (Giardia, Entoaemabeas) and blood parasites (Malaria). Entoaemebas can present with a dysentery which can cause fever. However, our patient did not have any diarrhoea. 

This area has a moderate risk of malaria, especially during the rainy season. Patients with malaria present with very unspecific symptoms: from fever with general malaise or headache and vomiting to seizures, coma and shock. Therefore, any patient with a fever in a tropical setting should prompt investigations for malaria. The most important element in the clinical diagnosis of malaria is a high index of suspicion. 

To reach the laboratory diagnosis, parasites should be seen or detected in blood. Blood film microscopy (thin and thick blood films) is the gold standard for malaria diagnosis, identifies the Plasmodium species and also quantifies the parasitaemia. However, in low resource settings, where microscopy is not always available or reliable, rapid diagnostic tests (RDT) are used to diagnose malaria. The RDTs detect Plasmodium antigens confirming the presence of parasites in the blood but don’t provide any information regarding the species or the parasitaemia. 

Patients with malaria can be classified into severe or non-severe malaria based on clinical and laboratory findings as per the WHO 2015 Malaria Guidelines. This classification is crucial as it will guide treatment. The most important complications of malaria infection in children are cerebral malaria, severe anaemia, respiratory distress due to acidosis and hypoglycaemia. 

Severe Malaria
Clinical findingsLaboratory
-Impaired consciousness/unrousable coma (Glasgow score <11, Blantyre score <3)- More than 2 convulsions in 24 hours- Prostration- Deep breathings/respiratory distress- Shock- Bleeding – Jaundice with parasitaemia > 2% – Severe anaemia with parasitaemia- Acidosis- Hypoglycaemia- Hyperparasitaemia- Haemoglobinuria- Renal impairment

Patients with severe malaria should receive parental antimalarial treatment with Artesunate and supportive management followed by a full course of oral artemisin combination therapy (ACT). Patients with non-severe malaria can be managed with oral antimalarial medication. 

On admission, the initial blood film was negative for haemo-parasites. Repeat blood films and a Malaria RDT (rapid diagnostic test which detects Plasmodium falciparum antigens in blood after 20 minutes) were requested. Repeat Blood film revealed presence of Plasmodium falciparum trophozoites with a parasitaemia of 2%. 

So the patient was diagnosed with Severe Malaria given the suspicion of CNS involvement and started on IV Artesunate. The patient had a good clinical response with resolution of fever and completed a course of oral Artemisin combination treatment (Artemeter Lumefantrine). However, after one week, the fever reappeared. This time, it was a low-grade fever with maximum peaks at 38.5. Repeat blood tests were normal. 

Malignancies: In this case, blood film did not reveal any blasts, chest XR was normal and abdominal ultrasound did not reveal any masses. BMA was not available locally and since the patient was otherwise well, this was not considered necessary. 


Connective tissue disorders: 

SOJIA, AR are very uncommon but still a differential diagnosis of persistent fever. In this setting, no resources were available for auto-antibodies testing, therefore clinical findings are the main way of diagnosing it. Since the patient did not have any rash, arthritis… this diagnosis was not considered. 

Acute Rheumatic Fever (ARF): this condition is quite common in low resource countries due to increased risk of streptococcal tonsillitis due to poor hygiene, overcrowding, poor accessibility to health facilities, fake drugs…  Acute rheumatic fever is an illness caused by an inflammatory reaction to streptococcal infection. It causes an acute, generalised inflammatory response. This illness targets specific parts of the body including the heart, joints, brain and skin. ARF typically leaves no lasting damage to the brain, joints or skin, but can cause persisting heart damage.  Our patient did not meet the Jones’ Criteria of ARF. 

Miscellanea (other possible causes of fever): 

Central origin fever: children affected with cerebral palsy or other neurological disorders relatively often present with chronic intermittent febrile episodes persisting for months. These episodes are not related to any infections but are actually arising from an abnormal thermal regulation resulting from the brain injury.

Hyperthermia from severe dystonia: children with cerebral palsy with dystonia can present with fevers and elevated basal temperature associated with elevated creatinine phosphokinase levels. 

Drug related fever: medications can trigger fevers. Common medications used in cerebral palsy are anticholinergic drugs (e.g. hyoscine) which can provoke unwanted fevers as a side effect. In addition, withdrawal of medications can present with fever (baclofen withdrawal syndrome). 

Lastly, factitious fever is a very challenging diagnosis. Sometimes admissions to hospital and close measurement of fevers plus observation of patient and carer interaction is as important as complementary tests. 

After 2 months of intermittent fever, it finally stopped. Basal temperature was always slightly elevated. Patient was diagnosed with central origin fever.

 The majority of PUOs are caused by: 

A: Malignancy

B: Connective Tissue Disorder 

C: Infections

D: Other diagnosis

E: Unknown diagnosis

The correct answer is C.

Infectious diseases are the main cause of PUO (about 38%), especially in younger children. No diagnosis is reached in 30% of cases but these tend to be benign and self-limited. This is followed by connective tissues disorders (13%) and Other diagnosis (13%). Lastly, malignancies are very uncommon but very important to consider given the severity of the disease.

A patient admitted to your hospital has been spiking fevers every day for 12 days. No other clinical findings are present. What is your next step? 

A: Repeat basic investigations, re-take clinical history, re-examine the patient, perform a Bone marrow aspirate.

B: Repeat basic investigations, re-take clinical history, re-examine the patient and do adequate imaging depending on clinical findings.

C: Perform a PET-CT to localise the pathology.

D: Perform autoimmune studies.

E: Perform a bone marrow aspirate.

The correct answer is B.

In many PUO cases, clinical findings are very subtle and can appear days after the fever. Therefore, re-taking the clinical history and re-examining the patient carefully is key to guide the complementary tests.

A Turkish 5-year-old girl presented with high fevers, profuse night sweating for 21 days.  Clinical detailed history revealed that parents are not consanguineous. She doesn’t have any relevant past medical history. She is fully vaccinated. The whole family was in Turkey for 2 months over the summer holidays. They were living in a farm in rural Turkey where they had goats, cows and chickens. They were drinking fresh milk from the cow. Based on the history, what diagnosis would you consider? 

A: Tuberculosis 

B: Bartonella (Cat-scratch)

C: Brucellosis

D: Toxoplasmosis

E: Lyme disease

The correct answer is B.

Brucellosis is a zoonotic infection caused by ingestion of unpasteurized milk from infected animals. It is also known as the Mediterranean fever. It is caused by a bacterium called Brucella melitensis. The main symptoms are fever, profuse sweating and joint and muscle pain. 

An unaccompanied asylum seeker from Uganda has just arrived in the UK. He refers to being a 12-year-old. He has had fevers for a prolonged time. On examination, he has splenomegaly. Blood tests revealed pancytopenia. Blood film is negative for malaria. HIV and hepatitis B, C negative. He said that in his country many people have these symptoms and they call it Kala-azar. What kind of tropical infection is he referring to? 

A: Visceral Leishmaniasis 

B: Schistosomiasis. 

C: Non falciparum malaria 

D: Visceral Larva Migrans

E: Echinoccocus granulosus

The correct answer is A.

Kala-azar is the local term for Leishmaniasis. This is a parasitic disease spread by the sand-fly. Main symptoms are fever, enlargement of spleen and liver and pancytopenia. Leishmaniasis is the second-largest parasitic killer in the world after malaria. Diagnosis is made by histological finding of amastigotes on spleen aspiration/bone marrow aspiration and RK39 Antigen detection.

A roadmap for fever of unknown origin in children- Rigante, D; Esposito S., International Journal of Immunopathology and Pharmacology. Vol.26 no 2, 315-326 (2013)

Fever in Children and Fever of Unknown Origin- Rajeshwar Dayal, Dipti Agarwal, Indian Journal of Paediatrics, 83 (1): 38-43 (2016)

Pyrexia of unknown origin-Mark Wood, Mario Abinun and Helen Foster. Archives of Disease in Childhood, Education and Practice, 89 ep 63-69 (2004) 

Barbi E, Marzuillo P, Neri E, Naviglio S, Krauss BS. Fever in Children: Pearls and Pitfalls. Children (Basel). 2017;4(9):81. Published 2017 Sep 1. doi:10.3390/children4090081

Antoon J,Peritz D, Parsons M., Skinner A.,Lohr J. Etiology and resource use of fever of unknown origin in Hospitalized children. Hospital Pediatrics, 8 (3).: 135-140(2018)

For malaria: 

https://apps.who.int/iris/bitstream/handle/10665/79317/9789241548526_eng.pdf;jsessionid=AD1DDC86455A8D51D25CFEEADF7E1C75?sequence=1

Website resources: 

https://pedemmorsels.com/fever-of-unknown-origin/

https://dontforgetthebubbles.com/ent-infections-immunodeficiency/

https://dontforgetthebubbles.com/recurrent-or-periodic-fevers-investigate-or-reassure/

https://dontforgetthebubbles.com/tuberculosis/

https://dontforgetthebubbles.com/claire-nourse-tuberculosis-at-dftb17/

https://radiopaedia.org/articles/tuberculous-cervical-lymphadenitis

https://gppaedstips.blogspot.com/search/label/Juvenile%20idiopathic%20arthritis

https://www.paediatricfoam.com/?s=kawasaki

https://gppaedstips.blogspot.com/search?q=kawasaki

https://dontforgetthebubbles.com/josh-francis-rheumatic-heart-disease-at-dftb17/



Please download our Facilitator and Learner guides

Febrile Child Module

Cite this article as:
Team DFTB. Febrile Child Module, Don't Forget the Bubbles, 2020. Available at:
https://doi.org/10.31440/DFTB.27356
TopicFebrile child
AuthorSarah Timmis
DurationUp to 2 hours
Equipment neededNone
  • Basics (10 mins)
  • Main session: (2 x 15 minute) case discussions covering the key points and evidence
  • Advanced session: (2 x 20 minutes) case discussions covering grey areas, diagnostic dilemmas; advanced management and escalation
  • Sim scenario (30-60 mins)
  • Quiz (10 mins)
  • Infographic sharing (5 mins): 5 take home learning points

We also recommend printing/sharing a copy of your local guideline.

The expectation is for the learners to have watched or read one of the basic links before the session.

What is the deal with fever? a good overview of the approach to a febrile child

NICE fever guidelines for kids • LITFL covers the NICE guidelines, plus a bit more

If you prefer to listen rather than read, there is a podcast that although long, is worth listening to and covers the approach to a febrile child: (1hr 14 minutes)

Pediatric Fever Without A Source

Fever is one of the most common presentations to the paediatric emergency department; it scares parents and it makes children miserable. So why does fever occur? A fever is a natural physiological response to infection. It occurs when either an exogenous (eg micro-organisms) or endogenous (eg TNF, interleukin-1 or 6) pyrogen is activated. These pyrogens, via a number of mechanisms, activate the anterior hypothalamus which ultimately results in an increase in body temperature (The pathophysiological basis and consequences of fever).  

This is crucial to understand – your body is in control of your temperature. This is not something an infection is doing to your body; it is something your body is doing to the infection. Of note- this is different from pathological hyperthermia, where your temperature is elevated by either hypothalamic dysfunction or external heat. These are extremely rare. (Hot Garbage: Mythbusting fever in children)

The process of having a fever is believed to be a beneficial response to an infection. The mechanisms by which a fever helps protect you from infection include:

  1. Higher temperatures inhibiting growth/replication of pathogens
  2. Higher temperatures promoting the immune response to infection
  3. It is also worth noting that bacteria are killed more easily by antibiotics at higher temperatures, so there is also a potential third mechanism.

With all this considered, it is not the presence of the fever that is the issue, but what the reason behind the fever is. This is what we, as clinicians, need to discern. First of all, is it infection (most likely in the paediatric population), if so, is this a serious infection? Or is the fever caused by something else (malignancy, drugs, autoimmune, endocrine)?

A father attends the ED with his 4 year child, who has a 2 day history of fever, his most recent temperature was 39.9oC and this has prompted his visit to the emergency department. The father describes his child as being otherwise well, but is extremely concerned about the height of the fever.

Describe how you would assess the child?

What investigations and treatment options would you consider?

You are happy with your assessment of the child, and would like to discharge him, however his temperature is 38.5oC. How do you proceed?

Is this child sick?

The Paediatric Assessment Triangle (from DFTB)

In some instances it will be fairly obvious if the child is unwell, they just ‘look unwell’. A tool that can help you put a system to this assessment is the paediatric assessment triangle. Which considers the childs: appearance, breathing and circulation. This will let you consolidate what you are worried about and allow you to communicate this to your colleagues. 

If all these appear to be in order, this is a reassuring sign. A happy child playing in the waiting room, whilst eating a packet of crisps is much less likely to be unwell with a serious bacterial infection than one that is quiet. Remember to write what you have observed in your notes. 

NG143 Traffic light tool (from NICE)

Once you have some observations you can also use the NICE traffic light table – which helps categorise children into green (well), amber or red (potentially unwell). If they score red, you know they need further workup, and potentially quickly. Green, then they can probably wait a bit to be seen. 

Take a full paediatric history, specifically asking about:


Normal self?
Eating and drinking?
Passing urine?
Bowels opening?
Drowsy?
Pulling at ears?
Vomiting?
Rash/ lumps and bumps?
Siblings, anyone else unwell?
Travel?
Immunisations?

This should also give you an idea about how worried the parents are, is it just the temperature, or is it something else? A high fever with a child who seems their normal self is far less concerning than a child with a normal temperature that just isn’t right.

This needs to be thorough, given that the majority of these kids will be discharged without further investigation. This means looking in ears and throats, looking at the skin hidden under clothing, looking at joints, feeling pulses. So undress the child. You may find a petechial rash, a lump, or more likely, some very enlarged tonsils. Get the child to walk if they are old enough, and stand on one leg and then the other. And when it comes to ears and throats get the parent on board and show them how to hold the child properly. 

Whilst you are hunting for the source, also note the absence of one- look for the signs of the scary infections, the petechiae, the reduced air entry on a lung base, the red knee.

Are you happy examining a child? https://vimeo.com/60599216 gives some top tips on how to examine different age groups

Also read https://dontforgetthebubbles.com/finding-fever/ for a step by step fever focussed examination guide

This step depends very much upon your assessment of the child. If you have found a source-treat that as appropriate. If full history and examination does not provide you with an answer, you have a fever with an unclear source. NICE helpfully has a set of guidelines for these: (NICE fever guidelines for kids • LITFL)

  • Investigate fever with no source if they have any red features –  this includes FBC, CRP, B/C and urine.  Consider LP, CXR, UEC and gas if indicated. 
  • Investigate fever with no source if there are any amber features unless deemed unnecessary by an experienced paediatrician. (this is the bit that could cause you to become unstuck, and you may want a senior to look over these)
  • Check urine for all children with fever (over 37.5) and no source, even if they are green (on the NICE traffic light systemt).

Consider the use of paracetamol or ibuprofen to bring down a high temperature in a hot and miserable child. If it makes the child feel better, it will make the examination process easier for everyone. NICE advises alternating antipyretics.

In many children with fever, the cause will be viral, the source of which may be obvious, or may still be unclear. If they are in a low risk group with a normal urine, they may be ok to go home with advice and a leaflet on the use of antipyretics, fluid management and safety netting advice. However as stated above these are only guidelines, if you are not happy you can always investigate, or admit for observation, and parents can always come back.

If the child has a fever but you have a well child that you have no concerns about then you do not have to wait for the temperature to come down before discharge. 

Give the parents advice on recognising red or amber signs by providing written information and/or arranging follow-up- most EDs will have a ‘fever’ leaflet to give to parents. 

Educating the parents about the nature of fever is important. Explain that “We treat fever with anti-pyretics because it makes the child feel bad, not because fever itself is bad.” Fever is due to a functional immune response. It is what is causing the fever that has the potential to do harm. As a result what the fever is, is not nearly as important as how the child looks or behaves. (The caveat being an under 6 month old where the height of fever is relevant)

On discharge tell them If the fever lasts for more than 5 days, the child should at least have a repeat physical exam by a clinician.

Finish with “But come back if you are worried about the child, even if you have only made it to the car park/ house/ doors of the ED”  
A good summary in video form on seeing a feverish child: https://rolobotrambles.com/listen-look-locate-an-approach-to-the-febrile-child-tipsfornewdocs/

A 5 week old girl has been brought in by her mother. Her mother reports the child seemed irritable so she took her temperature and it was 38.2oC. Pregnancy and birth was unremarkable and there have been no concerns since her birth. The child is feeding well and the history and examination are unremarkable, observations in the ED have been within normal limits, apart from her current temperature which is 38.5oC. Your initial assessment has not provided you with an obvious source for the infection.

When is a temperature classed as a fever?

How would you investigate this child?

How would you manage this child if they had a white cell count of 17 x109/L?

NICE consider >38oC to be a fever

RCEM considers a temperature of 37.5-38oC to be a low grade fever 

However, most people would agree that the difference between .1 of a degree is not significant, therefore infants with a temperature of 37.9 vs 38oC should be managed in the same way.

This child is under 3 months old

Any child with a fever >38oC that is under 3 months old is at ‘high risk’ of serious illness (‘red’ on NICE’s traffic light table NICE fever guidelines for kids • LITFL). If they have a history of fever, but none on assessment remember to ask about antipyretics.

According to NICE this child requires bloods (FBC, CRP, Blood cultures), a urine sample and if the history and exam suggests, a chest X-ray and/ or a stool culture.

A lumbar puncture should be considered and is indicated if the child is:

  • less than 1 month 
  • 1-3 months and unwell; 
  • or 1-3 months with WCC<5×109/L or >15×109/L.

The discussion here is if the child is ‘unwell’, or not. You have a few tools that can help you – the paediatric assessment triangle and the NICE traffic light table (referenced in the above case) can help you decide. However if in doubt, the child will be investigated, and you should be speaking to the paediatric seniors.


If this child had a WCC of 20 then this is an indication for IV antibiotics. 

IV antibiotics are required for children under the same criteria that a lumbar puncture is indicated: 

  • if less than 1 month; 
  • 1-3 months and unwell; 
  • or 1-3 months with WCC<5×109/L or >15×109/L.

The choice of antibiotic will come down to trust guidelines.

A 7 week old has been brought in by her mother because she felt very hot today, and has been ‘a bit grizzly’. Mum has given paracetamol and brought her to ED. Her temperature is 37.6oC on triage. On initial assessment you have no concerns and remaining observations are within normal limits. 

How should a temperature be taken? 

How would you investigate and manage this patient?

NICE has recommendations on this:

Do not routinely use the oral and rectal routes to measure the body temperature of children aged 0–5 years.

They advise in infants under 4 weeks: 

  • measure body temperature with an electronic thermometer in the axilla 

In children aged 4 weeks to 5 years use one of the following:-

• electronic thermometer in the axilla

• chemical dot thermometer in the axilla

• infra-red tympanic thermometer

It’s worth checking what your department uses and what the parent has been using. 
There are some small studies with low numbers of patients that suggest that layers of clothing can raise the skin temperature by up to 2.5°C with a minimal rise in rectal temperature in the very young (Feel the heat). Therefore undress children who seem inappropriately overdressed.

For this patient, guidelines are helpful, but they will not tell us what to do.

We know that 

1. Any child with a fever >38°C under 3 months old is a ‘red’ on NICE’s traffic light system, and this makes them at high risk of serious illness.

2. NICE guidelines suggest that the parents subjective perception of a fever should be considered valid and taken seriously by healthcare providers. 

There is a temptation to treat a child who is apyrexial in the department differently to one that does have a fever. Consider:

  • Has this child had an antipyretic? 
  • In the young, mums are usually right (There is a study from 1984 that shows in children under 2 yrs, mums were correct 90% of the time when they thought their child had a fever, although this dropped to 50% accuracy in over 2 year olds.)
  • Those with fever at home are equally at risk as those with fever in the department (A BMJ study reports that infants <60 days of age, with a history of documented fever are at equal risk for bacteraemia or meningitis as those with fever in the department. https://adc.bmj.com/content/103/7/665.)

So in summary, we have an infant with a normal temperature, who probably had a fever this morning. There are at least two ways of managing this, one is to treat as a fever which therefore means bloods (FBC, CRP, B/C), urine and if history suggests, a CXR and or stool culture. Given that there was parental concern this is probably the preferable option. The other is a period of observation to see how the child progresses, and see whether or not they spike a fever. 

Given that there are no clinical concerns at present, antibiotics prior to blood results are not indicated.

For a debate surrounding overtreating infants read https://dontforgetthebubbles.com/fever_under_60_days_of_age/

A 3 year old boy has returned to ED with a history of 6 days of fever, they have seen the GP twice, two and four days ago, and told it was a viral illness. However the fever is persistent and his parents are concerned. His past medical history includes two admissions for viral wheeze when he was younger, but is otherwise unremarkable. All immunisations are up to date, he goes to nursery and lives with his parents, he has no siblings but his mother is 9 weeks pregnant. On examination the child seems grumpy, he has a fever of 38.8 and a HR of 150 he has a rash across his face and torso and evidence of conjunctivitis.

You think the rash looks morbilliform, what are your concerns and how will you proceed?

What other differentials should you consider, and what examination findings would you be looking for?

How would you work this patient up?

Measles – A brief historical & clinical review

The MMR in the UK is given at 12 months and 3yrs 4 months, so this child will have had the first immunisations affording him 80-95% protection, https://em3.org.uk/foamed/15/7/2019/lightning-learning-measles. Measles therefore is unlikely but possible. Once he has had the second vaccination, this is quoted to afford 99% protection.  

Hopefully you are seeing this child in a side room, as measles can survive for up to 2 hours in air and is very contagious in the un-immunised population. 

It is likely wherever you are in the world, you will need to report this to your public health body. 

His mother is pregnant, check her vaccination status, if this is not complete and she has no history of disease, you need to advise her to see her GP ideally today, she may need a measles titre and, if this does not show previous exposure to the disease, human normal immunoglobulin (HNIG). You also need to enquire about other immunosuppressed/ non immunised contacts. 

A patient is infectious from 4 days before the onset of rash to 4 days afterwards, therefore he will need to be isolated until this period is up and nursery and other contacts need to be informed. 

Serum and saliva testing for measles is available.

Most children with measles can be discharged home

UK guidelines on managing measles exposure : Guidelines on Post-Exposure Prophylaxis for measles June 2019 

Poster: https://em3.org.uk/foamed/15/7/2019/lightning-learning-measles

Recurrent or Periodic Fevers – investigate or reassure? 

Think infection, inflammation or neoplastic. We know infection is common in paediatrics, and the other two are less so. The list of differentials is probably almost endless. There is a good article which lists a whole heap of causes of fever in children, and investigations which can be performed. 

However with this presentation, it is important to consider Kawasaki disease with this time scale of fever and measles. Other conditions worth considering are listed below: 

• Streptococcal disease (e.g. scarlet fever, toxic shock syndrome)

• Staphylococcal disease (e.g. scalded skin syndrome, toxic shock syndrome)

• Bilateral cervical lymphadenitis

• Leptospirosis and rickettsial diseases

• Stevens-Johnson syndrome and Toxic Epidermal Necrolysis

• Drug reactions

• Juvenile Chronic Arthritis

Kawasaki Disease 

You are looking for evidence of Kawasaki disease: The diagnosis is made on the basis of the following clinical criteria (A + B):

A. Fever ≥5 days

B. At least 4 of the 5 following physical examination findings:

  • 1.Bilateral, non-exudative conjunctivitis
  • 2.Oropharyngeal mucous membrane changes – pharyngeal erythema, red/cracked lips, and a strawberry tongue
  • 3.Cervical lymphadenopathy with at least one node >1.5 cm in diameter
  • 4.Peripheral extremity changes 
    • acute phase: diffuse erythema and swelling of the hands and feet
    • convalescent phase: periungual desquamation (weeks 2 to 3)
  • 5.A polymorphous generalised rash – Nonvesicular and nonbullous. There is no specific rash that is pathognomonic for KD

This child has had a fever for 6 days, is tachycardic and the source currently is unclear. It may be measles, however this is not clear cut. He is therefore not going home. Depending on other findings on examination he may also fit the criteria for Kawasaki disease he certainly needs bloods, FBC, U+E, LFTs, CRP, ESR, cultures and a urine dip. He does not require IV antibiotics at this point.

Kawasaki Disease the first 4 minutes covers the presentation and investigation of Kawasaki disease

Communication: Septic screen , taken from  Simulation Library, PaediatricFoam

Which of these is true, a 60 day old with a temperature of 38.5oC:

A: Fulfils the criteria for a lumbar puncture

B: Can be discharged without further investigation 

C: Needs IV antibiotics

D: Needs urine sent for urgent microscopy and culture

The correct answer is D.

This child will need further investigation, at the least bloods and serum cultures, however if they are well they may not necessarily need antibiotics or a lumbar puncture. All children under 3 months need urine sent, not dipped. Use dipstick testing for infants and children 3 months or older.

Which of these is false?

A: The height of the fever can make a difference to the how the child is managed

B: If a fever doesn’t reduce with an antipyretic the child needs admission to hospital

C: A 28 day old with a temperature of 38.5oC will need FBC, CRP and Blood cultures

D: It is recommended that children aged 4 weeks to 5 years have their temperature taken with an axillary probe or tympanic thermometer

The correct answer is B.

A is true because the height of the temperature does make a difference to the management of those under 6 months old

Presence of a fever, even one that does not reduced with an antipyretic is not an indication of a serious infection. It is perfectly acceptable to discharge a well child with a fever, with good safety netting. 

Which of these is true?

A: Kawasaki disease can be diagnosed with fever for > 5 days plus 3 of the B symptoms

B: Fever of over 39oC in a 3-6 month old automatically needs a full septic screen 

C: The higher the fever, the more likely it is to be a serious bacterial infection

D: Measles is infectious from 4 days before the onset of the rash to 4 days afterward

The correct answer is D.

Kawasaki disease is diagnosed with fever >5 days and 4 out of 5 B symptoms

A fever of >39 in a 3-6 month may need a full septic screen, the temperature alone would push them into NICE’s ‘amber’ category. However it depends on a few factors, including whether there is an obvious source and NICE recommends a review by an experienced paediatrician before performing a septic screen automatically on these patients. 

C is not true, there is no good consistent evidence to suggest a higher fever means a more serious infection 



Please download our Facilitator and Learner guides

Hot Garbage: Mythbusting fever in children

Cite this article as:
Alasdair Munro. Hot Garbage: Mythbusting fever in children, Don't Forget the Bubbles, 2020. Available at:
https://doi.org/10.31440/DFTB.22916

Juniper is a 3yr old girl brought in with her mother, with a 48hr history of fever. Her mum is particularly concerned because her fever was up to 39.8°C, didn’t come down with paracetamol and she describes an episode which sounds like a rigor. On examination, she has a temperature of 39.3°C, a runny nose and bright red tonsils, and looks otherwise well. You go to discharge her, but your colleague asks if you should wait to see if her temperature comes down with ibuprofen before sending her home?

 

Introduction

Febrile illnesses are the most common cause of presentation to acute paediatric medical services. This means that fever is the most common presenting symptom seen by paediatricians, and it is clearly a huge cause of concern for parents. Despite this fact, it is clear that in day-to-day practice that there is a widespread misunderstanding about fever, its purpose, and its clinical interpretation.

Well, no longer! Once you have finished reading, you will be a master of all things related to fevers in children. We will start with some basic understanding of the processes surrounding fever, and finish off with some mega myth-busting!

What is fever?

Fever is an elevated core body temperature, as part of a physiological response to infection regulated by the hypothalamus. This is crucial to understand – your body is in control of your temperature. This is not something an infection is doing to your body; it is something your body is doing to the infection. This is different from pathological hyperthermia, where your temperature is elevated by either hypothalamic dysfunction or external heat. These are extremely rare.

Note: there are other, non-infectious causes of fever, such as cancer, Kawasakis, and autoinflammatory conditions, but these are rare in comparison to infectious fever and are covered elsewhere.

 

What temperature counts as a fever?

At what threshold do we say a child has an elevated body temperature? This is more controversial than one might think, as actually the data from which we derive “normal” body temperature is extremely poor. The most common cut off for defining a fever is 38°C – but it is important to remember that there is nothing magic about 38°C compared to 37.9°C, and temperature is better taken in context or a trend, if possible.

How do we get fevers?

The process of developing fever is extremely complex, and our understanding is still developing. At present, our best explanation is that the process is triggered by the presence of chemicals referred to as pyrogens. Pyrogens can either be exogenous (such as parts of the microbe itself, like the lipopolysaccharide on the outside of bacteria), or endogenous, such as cytokines like IL1, TNF, Prostaglandin E2 and importantly IL6, which are released by immune cells when they detect an invader. These pyrogens act to increase body temperature peripherally, but importantly also trigger receptors in the preoptic nucleus in the brain. This releases PGE2 into the hypothalamus, which then sets a new target temperature. This target is met by many facets designed to increase heat, including:

  • Release of noradrenaline by the sympathetic nervous system, increasing thermogenesis in brown adipose tissue and causing peripheral vasoconstriction and piloerection (reducing heat loss)
  • Acetylcholine release stimulating muscle myocytes to induce shivering
  • Feeling cold”, inducing heat-seeking behaviours (warm clothes and blankets)

It is important to remember that the body is trying to get hotter. If you intervene with non-medicinal efforts to cool it down, it will work even harder to try to heat up.

Why do we get fevers?

The process of having a fever has been conserved across species from lizards to mammals, and even plants! This is because it is a beneficial response to an infection. The mechanisms by which a fever helps protect you from infection include:

  1. Higher temperatures inhibiting growth/replication of pathogens
  2. Higher temperatures promoting the immune response to infection

It is also worth noting that bacteria are killed more easily by antibiotics at higher temperatures, so there is also a potential third mechanism.

 

Summary

Fever is beneficial. When a pathogen causes infection, pyrogens stimulate the hypothalamus to increase the body temperature through several mechanisms, and this increased temperature helps inhibit the growth of the pathogen AND stimulates the immune system to fight it.

That was a lot of science. Don’t worry – it’s time to get clinical! All this science stuff is lovely, but what does this mean for our patients?

Clinical significance of fever

As we have ascertained, fever is beneficial. For this reason, when a child presents with fever, the fever itself is actually of no concern. What we are interested in is the reason for the fever. Is this fever the result of a benign, self-limiting, childhood infection – or is it associated with a serious bacterial infection? Trying to determine this is enough for its own blog article (the most important thing is the end of the bed assessment – see Andy Tagg’s excellent breakdown of the paediatric assessment triangle).

Ignore the fever itself – what’s important is ascertaining its cause.

Now, let’s get on and bust some myths that persist surrounding fever in children!

 

Myth 1 – Higher temperature indicates a serious infection

This is one of the most common concerns amongst parents. The particular height of temperature may be what prompts them to come to hospital, or even what prompts the health care provider to initiate more aggressive management or investigations.

The truth is that the relationship between the height of temperature and risk of serious illness is at best complicated, and at worst a dangerous distraction. There is a very poor correlation, with such woeful sensitivity and specificity that it will both grossly over and under-call serious infections (either if the high temperature is used to rule in, or lower temperature to rule out). The caveat to this is in younger infants (particularly under 60 or 90 days), who have a higher baseline risk of serious infections (and more to the point – once they spike a temperature will be managed aggressively regardless of how high it was). Some studies have shown an extremely weak association in older children, but not enough for it to have any meaningful influence on our management. A fever is a fever – higher temperatures should not be managed any differently than lower ones.

 

Myth 2 – Temperature not relieved by antipyretics indicates a serious infection

Another common misconception also linked to the myth above. Some fevers respond well to antipyretics, and some do not. We do not understand why this is the case, however, studies have not demonstrated that failure to respond to antipyretics is a useful indicator of a more serious infection. It is not very pleasant for the child to remain hot, but it does not mean they are at any higher risk. A child whose temperature does not respond to antipyretics should not be treated any differently to one that does.

Myth 3 – Rigors indicated a serious infection

This has been covered in-depth in a separate blog post – but to summarise; there is extremely weak evidence that rigors are associated with an increased risk of bacterial infection in children, which is irrelevant when factors that are more important are taken into account. There is also evidence of no increased risk. The presence or absence of rigors should not be a deciding factor in the management of febrile children.

Myth 4 – You must wait for a fever to come down before discharge

This may seem common practice for many of you working in acute paediatrics. If a child is febrile on arrival, people often want to wait to see the temperature come down before allowing them to be discharged (this should be differentiated from seeing observations normalize in the absence of fever – which is a more understandable if still slightly questionable practice). As we have seen, a fever merely indicates the presence of an infection. If you have ascertained the cause of the fever, or at least ruled out any red flags for serious causes, the ongoing presence or absence of a fever means nothing for the child. If it comes down before discharge, it will probably just go up again once they are home! There is no need to make them wait around for hours for no reason.

Myth 5 – Fever should be treated with antipyretics

We have established that fever is beneficial. Therefore, there is essentially no reason to treat a fever in and of itself. It will not cause harm, and it is probably helping. Some children tolerate having higher temperatures extremely well, so if they are playing happily or do not seem terribly bothered about their temperature of 39°C then you leave them well alone.

Treat the child, not the fever.

Myth 6 – Fever should not be treated with antipyretics

There is an opposing school of thought, which says that since fevers are beneficial, we should not treat them at all. Given how absolutely dreadful it can feel to have a fever (which many of us adults should be able to vouch for), many of us give medicines to try to bring the temperature down and make the child more comfortable. This is the right thing to do. Despite the potential benefits having a fever confers, there is no evidence of any clinically meaningful harms to treating temperatures in unwell children, or even in adults in ICU. If the child is distressed by the temperature, they should have antipyretics to make them feel more comfortable.

Summary

  • Fever helps your body to fight infection and is not dangerous (no matter how high)
  • The fever itself is not important. The cause of the fever is what matters
  • There is little to no evidence that higher temperatures, temperatures that don’t respond to antipyretics, or rigors indicate an increased risk of serious infection
  • Persisting fever on its own is not a reason to postpone discharge
  • Only treat fevers if they are causing distress. Treat the child, not the fever

 

Postscript: Febrile convulsions

When I posted my initial thread on twitter about fevers, there were many comments asking why I didn’t address febrile convulsions. This was mainly because these are worth a post to themselves (which they have here). In brief, febrile convulsions are extremely distressing for parents to observe, but they are common and they are very benign. We do not advise treating fevers to prevent febrile convulsions, and until recently, this was because there was no evidence that they had any effect in preventing them. A recent study from Japan did demonstrate a decrease in recurrence of febrile convulsions in children who had already had one if given regular PR paracetamol, however, there are major caveats to this study discussed in depth here.

 

For the more visual oriented, the talented Emma Buxton has created an infographic of the key reminders from this blog post:

The 32nd Bubble Wrap

Cite this article as:
Grace Leo. The 32nd Bubble Wrap, Don't Forget the Bubbles, 2019. Available at:
https://doi.org/10.31440/DFTB.20763

 

Article 1: Should we worry about fever after Meningococcal B immunisations?

Campbell G, Bland RM, Hendry SJ. Fever after meningococcal B immunisation: A case series. J Paediatr Child Health. 2019; 55: 932-937. doi:10.1111/jpc.14315

Why does it matter?

Meningococcal meningitis and septicaemia remain is one of the most serious bacterial infections (SBI). In Australia, the government subsidised vaccine schedule includes Meningococcal ACWY however Meningococcal B vaccine (Bexsero) can be purchased ($250-$500) for children 6 weeks to 11 years. It is an immunogenic vaccine, and paracetamol is recommended on the day of immunisation, but how can we be sure the fever is due to the infection and not something more sinister?

What’s it about?

A prospective case series from the ED at Royal Hospital for Children, Glasgow was performed on patients presenting between 2016-17. They identified 92 eligible infants under 3 months presenting with fever within 72 hours of Bexsero immunisations. The youngest infant was 7 weeks old.

Of these patients, 76 infants were discharged within 24 hours with majority undergoing at least one investigation (FBC, CRP, Urine MCS, NPA). Only 16 children of the 66 admitted remained in hospital for > 24 hours, with 12 undergoing an LP and completing 48 hours of IV antibiotics.

In this study, 26 children had an NPA performed with 12 positive for at least 1 virus, and one child represented with bronchiolitis.

Only one child in the cohort had SBI with an E.coli UTI. This infant also had a significantly elevated CRP and WCC compared with the other patients, and their fever started 54 hours after immunisation. The remainder had negative CSF, urine and blood cultures.

Clinically Relevant Bottom Line

Fever in the first 24 hours following the 2 month Meningococcal B vaccine is expected, and depending on the clinical exam and partial septic work up results, may be discharged home with reassurance. The key is to always be weary of the unwell looking infant and those whose fevers persist, as a full septic work up and IV antibiotics should be considered.

Reviewed by: Tina Abi Abdallah

 

Article 2:  Does iron fortified formula for Infants make a difference?

Gahagan S, et al. Randomized Controlled Trial of Iron-Fortified versus Low-Iron Infant Formula: Developmental Outcomes at 16 Years. The Journal of Pediatrics. 2019 June [epub] doi: 10.1016/j.jpeds.2019.05.030

Why does it matter?

Iron deficiency anaemia in infancy has long-term effects on the developing brain.  It is the most common nutrition disorder in the world.  Therefore, many countries routinely supplement infant formula with iron.  Recommendations for iron concentrations in infant formulas differ between guidelines, ranging from 4-12mg/L.Australian formulas generally contain between 6.7-9 mg/L.  There has been no study comparing the effects of iron-fortified formula vs low-iron formula on cognitive outcomes. 

What’s it about? 

Six-month old infants who did not have iron deficiency anaemia, were recruited from community clinics in Santiago, Chile.  They were randomised to iron-fortified (12mg/L) or low-iron (2.3mg/L) formula for 6 months and were followed-up at 16 years of age.  Of the 405 participants, those who randomised to iron-fortified formula (n=216) had lower scores than those randomised to low-iron formula (n=189) in 8 of the 9 tests.  Three of the 8 were statistically significant, and were in the domains of visual memory (p=0.02), arithmetic achievement (p=0.02) and reading comprehension achievement (p=0.02).  For visual motor integration, it was found that those with low haemoglobin at 6-months of age who received iron-fortified formula, outperformed those with low-iron formula.  The opposite was also true, with those with high haemoglobin at 6 months, receiving iron-fortification underperforming those with low-iron formula.  Animal studies have shown concern regarding the possibility of iron neurotoxicity in the growing infant, as well as the effects of iron exposure in early life on brain aging and neurodegenerative disease outcomes.

Clinically Relevant Bottom Line

This study from Chile suggests that  adolescents who received iron-fortified formula as infants from 6 to 12 months of age had poorer cognitive outcomes compared with those who received a low-iron formula. This could be related to iron neurotoxicity and there is a need for further studies to investigate the optimal level of iron supplementation in infancy.  Although on a public health level it may not be feasible, it may be ideal to individualise the optimal amount of iron for supplementation based on baseline haemoglobin or iron measures. 

Reviewed by: Lorraine Cheung

  1. American Academy of Pediatrics Committee on Nutrition recommends 10-12mg/L from birth. European Society of Pediatric Gastroenterology, Hepatology and Nutrition recommends 4-7 mg/L.

 

Article 3: Introducing paediatric procedural sedation in low-resource countries

Schultz, M & Niescierenko, M. Guidance for Implementing Pediatric Procedural Sedation in Resource-Limited Settings. Clinical Pediatric Emergency Medicine, 2019 In Press; doi: 10.1016/j.cpem.2019.06.004

Why does it matter?

Since its introduction, much research has demonstrated the safety and benefit of paediatric procedural sedation (when used with the proper monitoring). The benefits of procedural sedation include reduced procedure time and error rates; increased comfort of patients, parents and health care professionals; and reduced need for general anaesthesia for minor procedures. While paediatric procedural sedation is part of routine practice in high-income countries (HIC), it is almost non-existent in low- and middle-income countries (LMIC). The paper claims this is mainly due to a lack of skilled providers, not for lack of need. Providing proper clinical training to health providers in LMIC would help provide safe and adequate analgesia for children undergoing minor procedures.

What’s it about?

The authors of the paper have devised a paediatric procedural sedation curriculum, which was piloted at John F. Kennedy Hospital in Monrovia, Liberia. The pilot curriculum focuses solely on the use of ketamine, as it is cheap, widely available in Africa, has multiple routes of administration and is safe for use in children. The curriculum also allows for a single-practitioner method of procedural sedation, which is key in LMIC where there are limited number of health providers compared to the patient load. The curriculum is divided in three 2-hour sessions which consist of (1) introduction to procedural sedation, (2) resuscitation and management of adverse effect, and (3) monitoring and conclusion. All required teaching supplies were restricted to printed handouts, poster paper, markers and low-fidelity simulation equipment, thus eliminating the need for computers, software and electricity. Participants of this curriculum were 15 paediatric and surgical residents.

Clinically relevant bottom line

I was  fascinated with how the authors came up with this pilot curriculum for the Liberian hospital. Not only did they have to think about the costs of the individual piece of equipment in the sedation kit, but they also took into consideration the variable availability of electricity, the type of possible monitoring during sedation, and the scarcity of personnel. Too often we forget how lucky we are to have access to so many resources! The rollout of safe and routine paediatric procedural sedation is ongoing in Liberia and this is an initial step toward enabling safe procedural sedation for children living in LMIC.

Reviewed by: Jennifer Moon

 

Article 4: Supporting parents to CEASE smoking

Nabi-Burza E, Drehmer JE, Hipple Walters B, et al. Treating Parents for Tobacco Use in the Pediatric Setting: The Clinical Effort Against Secondhand Smoke Exposure Cluster Randomized Clinical Trial. JAMA Pediatr. Published online August 12, 2019. doi:10.1001/jamapediatrics.2019.2639

Why does it matter?

Exposure of children to secondhand and even thirdhand smoke (from toxins absorbed in clothing, carseats) is a serious public health issue. Smoke from cigarette smoke contains about 4000 chemicals, over 50 of which are known carcinogens. Second hand smoke increases the risk of children having SIDS, ear and respiratory infections, asthma exacerbations and teeth problems.

What’s it about?
The CEASE intervention (Clinical Effort Against Secondhand Smoke Exposure Cluster Randomized Clinical Trial) was developed between the AAP  and Massachusetts Tobacco Cessation and Prevention Program, and the Massachusetts General Hospital Center for Child and Adolescent Health Research and Policy. It focuses on 3 of the 5 As of tobacco cessation – Ask, Assist and Arrange Follow Up.

In this cluster RCT study run by the AAP, the CEASE intervention was delivered in paediatric clinics in 5 American states. The CEASE intervention included a smart tablet questionnaire, educational pamphlets and aids and training for staff to help screen for tobacco use and offer treatment to parents. Treatments were referral to a Quitline and/or provision of nicotine replacement therapy.
The study looked at the effectiveness and sustainability of this CEASE intervention 2 weeks and 2 years post start of intervention.

In a population of 8184 parents screened, 27.1% in the intervention group and 23.9% in the control group were smokers. Engagement in a treatment practice was 44.3% in the group vs 0.1% in the control. In the 2 year follow up of 9794 parents screened, 24.4% and   of the parents were smoking in the intervention and control practices respectively. There was a reduction in smoking prevalence in the intervention practices of 2.7% compared to an increase of 1.1% in the usual care control group. The NNT to treat to reduce one smoker was 27 individuals. For confirmed cessation (saliva tested, at least quit for 1 week), the NNT was 18.

The bottom line

We might take a smoking history routinely, but how often do we reach the next step of advising or assisting parents to quit? This trial shows that simple interventions can help improve uptake of treatment. Every parent who quits is one less child exposed to dangerous chemicals from tobacco smoke. The CEASE resources are freely available for use and adaptation https://www.massgeneral.org/children/cease-tobacco So why not move from contemplation to action and take the time to adopt a meaningful change in your own practice? There’s help on hand and it may be easier than you think to start!

Reviewed by: Grace Leo

 

If we have missed out on something useful or you think other articles are absolutely worth sharing, please add them in the comments! We are also looking to expand the Bubble Wrap team so please contact us if you’re interested in this! That’s it for this month. Many thanks to all of our reviewers who have taken the time to scour the literature so you don’t have to. 

Emerging infectious diseases : Mike Starr at DFTB18

Cite this article as:
Team DFTB. Emerging infectious diseases : Mike Starr at DFTB18, Don't Forget the Bubbles, 2019. Available at:
https://doi.org/10.31440/DFTB.20158

Mike Starr is not, despite what he tells you, the bassist for Alice in Chains. He is a general paediatrician and paediatric infectious diseases specialist at the Royal Children’s Hospital in Melbourne. He also happens to be a consultant in paediatric emergency medicine and plays a key role in the group that creates and collates the RCH clinical guidelines.