The Paediatric Papers are summaries of important journal articles for paediatric staff. They aim to provide concise, easily read clinical advice from high-impact publications.

Background

The Paediatric literature is constantly evolving and at such a rapid pace that it is difficult to keep up-to-date with the evidence base. Paediatric Papers aims to help by distilling the wisdom gained from journal articles into written, audio or multimedia summaries that can be consumed within two minutes.

Get involved

If you are interested in creating, or want to suggest, an article for a Paediatric Papers summary, please email education.hub@rch.org.au

A summary of how to make a Paediatric Papers summary can be found here.

Editors

Jye Gard, Amy Gray

Contributors

John Doan, Julian Dascalu, Jye Gard, Emer Ryan, Nelson Wang, Ary Sudarmana, Caitlin Richardson, Sarah Bush, Beth Cole

Article

McGuire K, Fung LK, Hagopian L, Vasa RA, Mahajan R, Bernal P, Silberman AE, Wolfe A, Coury DL, Hardan AY, Veenstra-VanderWeele J. Irritability and problem behavior in autism spectrum disorder: a practice pathway for pediatric primary care. Pediatrics. 2016 Feb 1;137(Supplement 2):S136-48.

How it helps

This consensus guideline, built upon a literature review, helps clinicians develop personalised management plans for children with Autism Spectrum Disorder (ASD) and irritable or other problem behaviours.

The article recommends a pathway of assessment, management, referral and initial treatment, to ensure all potential contributors to these difficult behaviours are addressed and optimised.

Want more…?

Study type

A multidisciplinary specialist group (child psychiatrists, behavioural psychologist and a developmental paediatrician from the U.S. Autism Speaks Autism Treatment Network Psychopharmacology Committee) performed a literature review of 5 common components in the assessment and treatment of difficult behaviours in children with ASD:

  1. Co-occurring medical conditions
  2. Co-occurring psychiatric conditions
  3. Lack of functional communication
  4. Psychosocial stressors, and
  5. Maladaptive reinforcement patterns

Following this review, the authors defined behaviours as either:

  1. Vocal or motor outbursts when angry (referred to as irritability), and
  2. Acts of aggression that can potentially cause harm either to themselves, others, or property (referred to as problem behaviours).

One arm of treatment for these behaviours is to use medicines, usually atypical neuroleptics, to try and help. However, medicines may cause additional problems including significant side-effects, and it is difficult to know when to initiate medicines and what to use.

This study acts to provide a step-by-step guide to helping children with these difficulties holistically - ensuring the overlay of all other issues are assessed and addressed before commencing medicines.

Major findings

The key tenets of the review are shown in the flowchart below. Overarchingly the authors describe the importance of:

  • Weighting care giver descriptions higher than those you make during the consultation (as it is an unfamiliar and non-routine setting likely to elicit atypical reactions)
  • Assess the impact these behaviours have on a child's life (learning, functioning, peer relationships, safety, and the safety of those around them)
  • Gather as much information as possible including corroborative observations and history from different care givers and settings. Are the behaviours situational?
  • Consider all possible contributors to these behaviours, including:
    • Medical problems (e.g. GI upset, seizures, dental pain, sleep difficulties)
    • Suboptimal communication strategies
    • Psychosocial stressors (e.g. bullying, parental stress, poor environment or opportunities)
    • Maladaptive behaviour loops (More here)
    • Comorbid psychiatric illnesses
  • Prioritise one specific behaviour to treat at a time
  • Regular re-evaluation every 3 months

Figure 1. A practice pathway

Where next

  • Moving forward new studies should:
    • Trial the efficacy of similar strategies for similar populations of children with ASD, in similar environments with similar difficulties. This is challenging as the overlay of possible contributors varies widely between different children.
    • We know the benefits and adverse effects of certain medicines for children with similar difficulties can vary widely. Is this an environmental effect? Is there a role for personalised medicine?

Additional resources

  • Read the original article for Table 1 – a detailed questionnaire to help in assessing children and their behaviours.
  • Read more about the medicines often trialled to help reduce the impacts these behaviours have on a child's daily function .
  • What do the authors mean by maladaptive behaviour loops.

Summary by Jye Gard, August 2021

Article

Twilhaar ES, de Kieviet JF, van Elburg RM, Oosterlaan J. Academic trajectories of very preterm born children at school age. Archives of Disease in Childhood – Fetal and Neonatal Edition. 2019;104F419-F423

How it helps

This is the first longitudinal analysis of how very preterm (VP) children perform academically during their primary school years compared to their peers born at term. VP children frequently have problems in arithmetic, reading comprehension and spelling that persist throughout primary school. However, they show a similar rate of academic progression to term-born children, which suggests that with early intervention VP children can 'catch up' to their peers. 

Want more…?

Study type

A longitudinal analysis of Dutch children using a national educational measurement (in Amsterdam, primary school children undergo a variety of standardised tests over grades 1 - 6).  

A selection of VP children (GA <32/40, many of whom also had very low birth weights <1500gthat had participated in a different study whilst cared for in an Amsterdam NICU from 2001- 2003 were recruited. 52/102 eligible children agreed to participate in the new study. 

58 classmates and students from neighbouring schools who were born at term and free of developmental, behavioural and learning disorderswere used as study controls. All children were tested in arithmetic, spelling and reading comprehension at multiple points between grades 1 - 6.  

Major findings

  • VP children scored lower in all three subjects over the course of their primary school education. 0.53 SD lower in arithmetic, 0.31 SD lower in reading comprehension and 0.21 SD lower in spelling. i.e. Between 2 to 5 letter grades lower than their peers. 
  • One in three VP children repeated a grade. 
  • VP children showed the same academic growth over time compared to controls though were generally unable to reach the same performance levels by the end of the study. i.e. even with early intervention, six years was too short for VP children to achieve grades similar to the ex-term peers 
  • VP children were more likely to require educational assistance and have a lower education level at the end of study  

Where next

  • Further studies are needed to investigate which learning skills and neurocognitive functions can be targeted to utilise the learning potential in VP children to allow them to reach similar education levels as their term-born peers 

Additional resources

Summary by Aryanto Sudarmana, August 2020

Article

Tagg A, Roland D, Leo GS, Knight K, Goldstein H, Davis T, Don't Forget The Bubbles. Everything is awesome: Don't forget the Lego. Journal of paediatrics and child health. 2019 Aug;55(8):921-3.

How it helps

Parents can be reassured that certain small objects, like a Lego head, if ingested - will pass through a regular, gastrointestinal tract without issue. While assessed in fully grown adult doctors, the authors suggest the results can be extrapolated to be relevant to a paediatric population.

Want more…?

Study type

This is the world’s first cohort study assessing the time between pop (ingestion) and drop (faecal elimination) for a regulation Lego head.

Six health care professionals working in paediatric hospital care across two continents were included in the study. Exclusion criteria included previous GI surgery, inability to ingest foreign objects and aversion to searching through faecal matter. No subjects were excluded. One subject was unable to locate the Lego head.

Major findings

  • The FART score (“Found and Retrieved Time”) ranged between 1.14-3.04 days with an average of 1.71 days
  • SHAT scores (“Stool Hardness and Transit”) were calculated by: Sum of Bristol Stool Chart scores over a time period/time period in days. No difference in SHAT scores was observed pre- and post-ingestion suggesting that the Lego heads did not have an impact on the consistency of bowel motions
  • There was no association between the SHAT score and FART score
  • The observed FART scores were lower than the equivalent time observed for passage of ingested coins in a previous study
  • The authors conclude that material that can pass via the pyloric sphincter will pass through the anal sphincter

Where next

  • The authors suggest further studies to assess why coins and Lego heads differ in transit time – their proposed design includes three arms: ingestion of coin, ingestion of Lego head and, ingestion of Lego figurine holding a coin.
  • Whilst most ingested small objects can be passed without issue, exceptions include:
    • Button batteries
    • Superabsorbent polymers that can cause impaction (e.g. craft supplies)
    • Two magnets or a magnet and a metal object
    • Lead based objects – that can cause poisoning
    • Objects with multiple parts that can break apart

Additional resources

Summary by Ary Sudarmana, March 2021

Article

Wang ME, Biondi EA, McCulloh RJ, Garber MD, Natt BC, Lucas BP, Schroeder AR. Testing for meningitis in febrile well-appearing young infants with a positive urinalysis. Pediatrics. 2019 Sep 1;144(3).

How it helps

In America, there remains wide variation in practice for which infants with a positive urinalysis undergo LP. Infants <30 days old with raised inflammatory markers are more likely to have an LP. Hospitals who see lots of febrile infants (>10 cases of fever of unknown origin per day) are more likely to do LPs. Infants >30 days old with a positive urinalysis who "appear well" do not need LP.

Want more…?

Study type

The researchers performed a secondary retrospective analysis on data that had already been collected and used for a different quality improvement project (Reducing Excessive Variability in Infant Sepsis Evaluation - REVISE). The data included 20,570 "well looking" infants (7 - 60 days old) that attended one of 124 American hospitals between September 2015 to November 2017 for fever (T >38oC) of unknown origin. Infants who were "unwell looking" or at increased risk of bacterial infection were excluded such as children:

  • Assessed as either: toxic, ill-appearing, lethargic, sick-appearing
  • Comorbidities including genetic, neuromuscular and developmental disease
  • Diagnosed as having bronchiolitis
  • So unwell that they needed to be transferred to different hospital
So unwell that they needed to be transferred to different hospital

Major findings

  • Positive urinalysis = dipstick positive (including trace) for either leucocyte esterase or nitrates, or urinalysis showed >5 WCC per high-power field
  • Abnormal inflammatory markers = >1 of WCC <5000 or >15000m3, absolute band count >1500 cells mm3 per band-to-neutrophil count >0.2, elevated CRP or procalcitonin
  • 9/10 had urinalysis before discharge from ED. Of these, 2/10 had positive urinalyses. 90% had collection either via SPA or in-out catheter
  • 70% of children with positive urinalyses underwent LP. Most children who underwent LP were started on empirical antibiotics, regardless of a positive or negative urinalysis.
  • Infants <30 days with abnormal inflammatory markers and a positive urinalysis who attended a hospital that saw >10 infants with PUO each day, were more than 4 times more likely to have an LP
  • Of the 1061 infants with positive urinalysis who did not have LP, 70 received a 'full course' of antibiotics
  • There were no cases of delayed diagnosis of meningitis (representation to the same hospital within 7 days of discharge from ED or inpatient admission)
  • No infants >30 days old who had a positive urinalysis but no LP, returned to the same hospital within 7 days of discharge

Where next

  • The study builds on previous research regarding the prevalence of concomitant meningitis in children with UTIs - found to be 0.8 - 1.2% in the first month of life, 0 - 0.3% in the second
  • Are practitioners really that good at selecting out patients who are at higher risk of meningitis and require LP?
  • Care should be taken with interpretation of the results:
    • Infants may have presented to different institution
    • LPs may have been declined by the family or unsuccessful (9 infants without LP had stays >14 days but we don't get more details about them)
    • Short course of antibiotics may have been enough to treat undiagnosed meningitis

Additional resources

Summary by Jye Gard, June 2020

Article

Lyttle MD, Rainford N, Gamble C, Messahel S, Humphreys A, Hickey H, Woolfall, K, Roper L, Noblet J, Lee, ED, Potter S, Tate P, Iyer A, Evans V, Appleton R E. Levetiracetam versus phenytoin for second-line treatment of paediatric convulsive status epilepticus (EcLiPSE): a multicentre, open-label, randomised trial. The Lancet. 2019 May; 393: 2125-34.

How it helps

  • Convulsive status epilepticus (CSE) is the most common paediatric neurological emergency worldwide. The length of CSE is inversely proportional to a child’s risk of associated neurological morbidity.
  • The long-standing recommendation for Phenytoin as the second-line choice for CSE management after administration of Benzodiazepines has had limited trial evidence and significant safety concerns, including risk of cardiovascular compromise.
  • This trial is one of the largest comparing Levetiracetam to Phenytoin head-to-head.

Want more…?

Study type

An open-label randomised control trial across 30 UK emergency departments, which took place between July 2015 & April 2018. Children between 6 months to 18 years with CSE requiring second line treatment were randomised to receive either Phenytoin (20mg/kg IV or IO over 20 minutes) or Levetiracetam (40mg/kg IV or IO over 5-10 minutes). Children were recruited following deferred consent. The study’s primary outcome was looking at the time from randomisation to clinician assessment of seizure termination.

Major findings

The trial showed that Levetiracetam is as effective as Phenytoin in achieving cessation of CSE. Although the study showed adverse effects in both treatment arms, Levetiracetam had less serious adverse effects (ie no significant hypotension, decreased conscious state) associated with administration. There is also the added benefit of a quicker administration time with Levetiracetam. Due to these factors it should be considered an appropriate second line anticonvulsant for management of CSE in the emergency department.

Where next

This study was not powered to assess adverse drug reactions. Further studies or systematic reviews should assess in greater detail the adverse events associated with second-line anticonvulsants in CSE management.

Additional resources

Summary by Beth Cole September 2020

Article

Pierce MC, Kaczor K, Aldridge S, O'Flynn J, Lorenz DJ. Bruising characteristics discriminating physical child abuse from accidental trauma. Pediatrics. 2010 Jan 1;125(1):67-74.

How it helps

This study provides a clinical decision rule to help identify children ≤4 years old who are at high risk for physical abuse. The tool uses a child’s age, where the bruises are located and how many there are, to help discriminate between bruising typical of abusive versus accidental trauma.

Want more…?

Study type

A retrospective case-control single-site study of patients who were admitted to the Chicago Children’s Memorial Hospital’s Paediatric Intensive Care Unit (PICU) with injuries during a 3 year period. All children ≤4 years old who were admitted to the PICU who had an injury identified as due to abuse or an accident, were included. Additionally any child <1 year old was included (regardless of whether a cause was identified). Children with injuries from an indeterminate cause or who had underlying coagulation disorders (e.g. haemophilia, cancers) were excluded. The study used a variety of criteria to categorise the cause of an injury as either accidental or abuse (see Where Next).

Major findings

Where next

  • Studies into child abuse are difficult for a multitude of reasons including: ethical challenges, trickiness of consent, problems with verifying the validity of caregiver and child reports
  • The study group used five criteria to categorise injuries as likely due to abuse vs an accident. These were
    • To ‘rule-in’ abuse, subjects had to either:
      • Have injuries as identified by their local trauma registry (similar to Victoria’s) as “abuse”
      • Have the hospital team determine the injuries to be highly suggestive of abuse
      • The stated cause of injury (whom was not stated) did not account for the type, severity and/or number of injuries
      • History of another cause (e.g. trauma) was absent, vague or changing
      • State social services determined that the child was abused
To ‘rule-out’ abuse and label the injuries as accidental, subjects had to either
      • Have injuries identified by their local trauma registry as an “accident”
      • The hospital team had no concerns of abuse
      • The stated cause of injury was consistent with the injury (or injuries)
      • The history provided was detailed, thorough and consistent
      • No indicators of abuse were found on skeletal survey, forensic tam or social service evaluations
  • What do you think of these criteria?

Additional resources

  • Learn more about tools used to help investigate and classify injuries on the VFPMS website
  • Listen to our podcasts on how to take a forensic history and speak with families about forensic concerns
  • Read more about the evolution of this clinical decision tool here (Pierce MC, Kaczor K, Lorenz DJ, Bertocci G, Fingarson AK, Makoroff K, Berger RP, Bennett B, Magana J, Staley S, Ramaiah V. Validation of a clinical decision rule to predict abuse in young children based on bruising characteristics. JAMA network open. 2021 Apr 1;4(4):e215832-.)
Summary by Jye Gard, July 2021

View the infographic for this topic

Article

Pfeiffer H, Crowe L, Kemp AM, Cowley LE, Smith AS, Babl FE. Clinical prediction rules for abusive head trauma: a systematic review. Archives of disease in childhood. 2018 Aug 1;103(8):776-83.

How it helps

This systematic review provides clinical prediction rules to help identify children in the emergency department, on the wards and in ICU who are at high risk of having an inflicted head injury.

Clinical red flags may present differently in different settings and at different time points in clinical care. This study helps tease out which clinical clues can help discriminate between head injuries typical of accidental trauma versus possible abusive trauma in different contexts.

Want more…?

Study type

A systematic review of 20 years of medical literature (January 2016 to August 2016) using 3 databases found 110 articles on how to detect abusive head trauma in children.

Only three studies met the study’s inclusion criteria of:

  • High methodological quality as per a scoring system developed in a previous systematic review by Maguire et al
  • Had key elements in keeping with a high-quality clinical prediction tool, including the use of at least three variables in helping guide risk analysis
  • Specifically analysed head trauma (rather than the detection of child abuse in general).

The three studies analysed different clinical variables in different clinical settings. Each developed their own clinical prediction rules (CPredRs).

The systematic review compared and tried to amalgamate the findings from the heterogeneous studies - were there red flags to look out for in every setting? Was one study’s CPredRs superior and more reliable in detecting inflicted head injuries?

Major findings

Comparison was tricky as the three studies differed in many ways:

  • Primary focus of the CPredR
    • Pediatric Brain Injury Research Network tool: To prompt intensivists to do additional screening for abusive head injuries
    • Predicting abuse head trauma tool: To prompt clinicians to seek the help of a forensic specialists
    • Pittsburgh Infant Brain Injury Score: To prompt emergency physicians to do neuroimaging as part of their investigation panel
  • Population studied (e.g. ages, sites, how unwell)
  • Types of data used (retrospective vs prospective, single vs multisite)
  • The time point along the diagnostic journey where they are to be used
  • The weight each CPredR placed on certain positive findings (e.g. how significant certain elements of the history were rated)
  • How they classifiedconfirmed abusive head trauma” (e.g. from a multidisciplinary assessment, court proceedings)
  • How they classifiedconfirmed non-abusive head trauma” (e.g. not every child had an MRI brain to check)

The variables common to all 3 studies were cutaneous injuries and respiratory compromise (e.g. apnoea).

Other variables that were highlighted as significant included:

  • Certain skull fractures (multiple, comminuted, bilateral, diastatic, non-parietal, non-linear)
  • Types of ICH (bilateral or interhemispheric subdural)
  • Seizures
  • Retinal haemorrhages
  • Rib fractures
  • Long-bone fractures
  • Head circumference >85th percentile
  • Low haemoglobin (<113g/L)

Where next

Studies into child abuse are difficult for lots of reasons including: ethical challenges, trickiness of consent, problems with verifying the validity of caregiver and child reports.

Moving forward new studies:

  • Must test how valid these studies are in different settings (e.g. can the Pittsburgh Infant Brain Injury Score, developed to prompt ED clinicians to do an MRB, also be used to help ward clinicians to decide on the same?)
  • See if the rules complement each other when used together. Is using more than one CPredR helpful in getting clear on the need for further assessments?

Additional resources

Summary by Jye Gard, August 2021

Article

Parker CM, Cooper MN. Prednisolone versus dexamethasone for croup: a randomized controlled trial. Pediatrics. 2019 Sep 1; 144(3):e20183772.

How it helps

Use steroids in children with croup and stridor at rest to reduce the severity of symptoms, risk of a readmission or ICU admission. Steroids also decrease the length of hospital and emergency department stays. The type of steroid doesn't matter.

Want more…?

Study type

Prospective, double-blind, non-inferiority randomised controlled trial at a single site (two hospitals linked to a tertiary paediatric emergency department Perth).

Major findings

  • A single dose of high-dose dexamethasone (0.6mg/kg/dose), low-dose dexamethasone (<0.15mg/kg/dose) and prednisolone (1mg/kg/dose) were found to be equally efficacious in reducing hospital representations and symptom severity 1-hour post drug administration (Westley Croup Score).
  • Children with prednisolone were more likely to be given a second dose on re-presentation.
  • Some children given low-dose dexamethasone had poor symptoms scores at 3 hours. There may be a small population where low-dose dexamethasone has a delayed effect.

Where next

  • The study looked at non-inferiority, not superiority. New studies are needed to determine which steroid is best.
  • The study population was limited to one site and did not include atypical croup presentations (children <6 months, >20kgs).

Additional resources

  • Learn more about the Westley Croup Score here

Summary by Jye Gard, April 2020

Article

Peters RL, Koplin JJ, Gurrin LC, Dharmage SC, Wake M, Ponsonby AL, et al. The prevalence of food allergy and other allergic diseases in early childhood in a population-based study: HealthNuts age 4-year follow-up. The Journal of allergy and clinical immunology. 2017; 140(1):145-53.e8.

How it helps

The HealthNuts group provide the first longitudinal estimate of food allergy prevalence in Australian infants.

Children may ‘grow out’ of previously confirmed challenge-confirmed food allergies – they are more prevalent in 1 year old than 4 year olds.

Prevalence of allergy among children in Melbourne, Australia is remarkably high compared to the rest of the world.

Want more…?

Study type

Population-based, longitudinal cohort study of children in Melbourne, Australia reporting prevalence of food allergy (egg, peanut, sesame). Initial prevalence data had been described previously in 5,276 12-month-old infants (Allen 2015). Follow-up data was collected from the participants at 4 years of age.

Challenge-confirmed food allergy was defined by:

  • Positive test of sensitisation (skin prick testing wheal above cut-off for age and/or elevated specific IgE ³35 kU/L), and
  • Positive oral food challenge.

Major findings

  • Prevalence of challenge-confirmed food allergy was less at 4 years of age (3.8%) compared to 12 months of age (11.0%).
  • Between 12 months and 4 years of age, the most marked decrease in prevalence was seen with egg allergy (see below).
  • 5% of children experienced symptoms of an allergic disease (food allergy, asthma, eczema and/or allergic rhinitis) in the first 4 years of life.
  • International data reports much lower prevalence of food allergy than described by HealthNuts.
Prevalence of Allergic Outcomes
  1 year prevalence % (95% CI) 4 years prevalence % (95% CI)
Peanut allergy 3.1 (2.7-3.6) 1.9 (1.6-2.3)
Egg allergy 9.5 (8.7-10.3) 1.2 (0.9-1.6)
Sesame allergy 0.6 (0.5-0.9 0.4 (0.3-0.6)
Any challenge-confirmed food allergy 11.0 (10.1-11.9) 3.8 (3.3-4.4)

 

Where next

  • Some participants from the original cohort were lost to follow-up at 4 years of age (18.7%), and others declined to undergo a food challenge at 4 years.
  • Not all allergens were tested in the follow-up period, e.g. IgE-mediated cow’s milk allergy.
  • Further follow-up is planned for the HealthNuts participants at 6 and 10 years of age, with repeat questionnaires and formal allergy testing.

Additional resources

Summary by John Doan, May 2020

Article

Mitre E, Susi A, Kropp LE, Schwartz DJ, Gorman GH, Nylund CM. Association between use of acid-suppressive medications and antibiotics during infancy and allergic diseases in early childhood. JAMA pediatrics. 2018 Jun 1;172(6):e180315-.

How it helps

Infants (<6months) who are prescribed acid-suppressant medicines are more likely to be prescribed antibiotics before they turn 12 years old. Infants who are prescribed acid-suppressant medicines are more likely to develop allergies. Even if they only take a short course (<60 days). Children (<12 years old) who are prescribed antibiotics are more likely to develop allergies.

Want more…?

Study type

A retrospective cohort study of 792, 130 children of parents who work or worked for the United States Department of Defence. The researchers used the Military Health System (MHS); defence workers and their families are provided with subsidised healthcare and the MHS is a database that records their medical billing data (inpatient and outpatient). Akin to our PBS. The study expressed most of their findings using Hazard Ratios.

Major findings

  • Infants who received acid-suppressants (either H2RAs or PPIs) have significantly higher risk of every type of allergic disease. Infants given PPIs a little more so than those given H2RAs.
  • The risk is dose-dependent. Infants who took long courses of acid-suppressants (>60 days) were at higher risk, than those who had shorter courses. But even children who had short courses had significantly increased risk!
  • The most common allergic disease children given acid-suppressants as infants developed was food allergies. Especially cow’s milk allergy (their risk increased from 2/100 to 4/100 – a 52% jump in risk). There was also a marked increase in their risk of childhood asthma, allergic conjunctivitis, and urticaria.
  • Infants given antibiotic also had an increased risk of all allergic diseases, especially asthma (a greater than 2-fold risk!). There was also a marked increase in their risk of food allergies (cow’s milk and egg), anaphylaxis, allergic conjunctivitis and medication. As well as atopic dermatitis, allergic rhinitis, contact dermatitis and urticarial.
  • Unlike acid-suppressive medications, antibiotics did not demonstrate a dose-dependent risk for the development of allergies.
  • Children who were prescribed antibiotics during infancy, were more likely to also be prescribed acid-suppressants. And vice-versa.
  • The study recorded interesting data about how US doctor prescribing preferences differ from ours

Where next

  • Are there prescriber biases at play? If so, what are they?
  • Is there other protective or risk factors to consider? Is it safer to prescribe these medicines to certain cohorts?
  • Can the risk be mitigated?

Additional resources

Summary by Jye Gard, May 2020

Article

Leigh J, Rickard M, Sanger S, Petropoulos J, Braga LH, Chanchlani R. Antibiotic prophylaxis for prevention of urinary tract infections in the first year of life in children with vesicoureteral reflux diagnosed in the workup of antenatal hydronephrosis: a systematic review. Pediatric Nephrology. 2020 Apr 30:1-8.

How it helps

Currently there is no strong evidence for either the use or disuse of prophylactic daily antibiotics for children <1y found to have vesicoureteric reflux

Want more…?

Study type

A literature review (18 studies were included from 6895 found across four databases) where the researchers used software to exclude irrelevant articles based on their titles and abstracts.  Two researchers then read the full texts of the remaining articles (or 3 in a tie-break) to decide if they met the inclusion criteria.

Major findings

  • Daily prophylactic antibiotics are often empirically prescribed for infants with vesicoureteric reflux (VUR) with the aim of reducing their rate of urinary tract infections (UTIs) in the first year of life
  • Whilst there is some evidence that this may hold true for infants with symptomatic VUR (e.g. those who have already had a UTI), there is a paucity of evidence to start them for infants with asymptomatic VUR (i.e. as prophylaxis for infants who have yet to have a UTI)
  • The authors intended to perform a meta-analysis to test the hypothesis that infants with asymptomatic VUR would benefit from prophylactic antibiotics.
  • However, what they actually found was that the meta-analysis could not be completed – the current pool of available studies are of low-quality, retrospective and most do not include a comparison control group (i.e. researchers only report on the findings of children who took prophylactic antibiotics). Those that did used tiny sample sizes.
  • Without good data, very little could be concluded. There was some evidence that 15% of infants on prophylactic antibiotics will still develop at least one UTI anyway. Most were female.

Where next

  • This is a good example of the role of a literature review – to help identify a gap in research

Additional resources

  • Learn more about research tools like
    • Guidelines for how to conduct a systematic review or meta-analysis PRISMA (Preferred Reporting Items for Systematic Review and Meta-Analysis)
    • Guidelines on how to assess non-randomised studies (The Newcastle-Ottawa Scale)
    • Software you can use to help with Cochrane style reviews
  • The benefit of prophylactic antibiotics in infant with symptomatic vesicoureteric reflux

Summary by Jye Gard, May 2020



Article

Whittaker E, Bamford A, Kenny J, Kaforou M, Jones CE, Shah P, Ramnarayan P, Fraisse A, Miller O, Davies P, Kucera F. Clinical Characteristics of 58 Children With a Pediatric Inflammatory Multisystem Syndrome Temporally Associated With SARS-CoV-2. JAMA. 2020 Jun 8.

How it helps

Children with evidence of previous SARS-CoV-2 infection were found to develop signs and symptoms of Kawasaki’s disease (KD), including: Fever for 3 or more days, mucocutaneous signs (rash, conjunctival injection), high inflammatory markers, cardiogenic and/or distributive shock ("KD shock syndrome") and coronary aneurysms.

Want more…?

Study type

A case series of 58 children from 8 different hospitals in England who:
  1. Met criteria for Pediatric Inflammatory Multisystem Syndrome temporarily associated with SARS-CoV-2 (PIMS-TS)
AND
  1. Either SARS-CoV 2 PCR or IgG positive (using Epitope Diagnostic Inc ELISA kits)
The clinical symptoms, signs, laboratory and imaging results of children with PIMS-TS were compared to those of children who suffered from other paediatric inflammatory disorders such as KD, KD shock syndrome and toxic shock syndrome with the aim to help clinicians "pattern recognise" PIMS-TS and distinguish it from similar illnesses.

Major findings

Clinical features
  • Children were investigated for PIMS-TS if they had fevers that persisted for at least 3 days. Some untreated patients had fevers that persisted for as long as 19 days!
  • The next most common features were abdominal pain (53%), erythematous rash (52%), conjunctival injection (45%) and red cracked lips (29%). Rarer features were lymphadenopathy (16%), headache (26%), swollen hands and feet (16%), sore throat (10%)
  • Half of the patient were admitted to ICU for at least one of either inotropic support (47%), mechanical ventilation (43%) or treatment of acute kidney injury (22%)
Laboratory features
  • Most (40/46 children tested) were IgG positive, fewer (15/58) were PCR positive. It didn't matter whether children tested IgG or PCR positive, the clinical features were the same.
  • Many had very high inflammatory markers including CRP >220mg/L, Neutrophilia >13x109/L and Ferritin >610µg/L
  • Two thirds also had raised troponin levels and most (83%) has raised pro-B-type-BNP.
  • Only 2 patients had evidence of a new second virus (both adenovirus and enterovirus, EBV)
Clinical course features
  • Children fit one of three clinical patterns
    1. Persistent fever and raised inflammatory markers only
    2. Shock with left ventricular dysfunction, raised cardiac markers +/- dysrhythmias
    3. Kawasaki's disease or atypical Kawasaki’s disease
  • Children from all three groups developed coronary aneurysms!
How to "pick" a PIMS-TS case
  • Blood count: Compared to 'regular' KD cases, children with PIMS-TS had a higher WCC, neutrophilia, CRP and more profound anaemias and lymphopenias. Children with PIMS-TS also had mild thrombocytopaenia (unlike in KD where we often see thrombocytosis).
  • Biochemistry: higher troponins, fibrinogen levels.
  • Age: PIMS-TS children tended to be older

Where next

  • Prospective trials/data collection
  • Standardised methods for detecting SARS-CoV-2 so like can be compared to like

Additional resources

  • Read the World Health Organisation (WHO) definition of PIMS-TS here
  • Check out a similar study done in New York
  • A helpful image of the features of PIMS-TS from Don't Forget The Bubbles
Summary by Jye Gard, June 2020

Article

Ibrahim L, Babl, F, Hopper S, et al Cellulitis: Oral versus intravenous and homer versus hospital-what makes clinicians decide Archives of Disease in Childhood. 2020;105: 413-415

How it helps

This study describes physicians embedded decision-making rules about route and locations of antimicrobial therapy for cellulitis. The purpose of this study was to understand barriers to changing practice.

Want more…?

Study type

Web-based anonymous staff survey including a clinical scenario.

Major findings

  • There was a wide variation in perceived risk of bacteraemia in the clinical scenario.
  • There was a 50:50 split in terms of preference for treating intravenously as either an inpatient or outpatient.
  • Lymphangitis, fever, having already received 24hrs of oral antibiotics, and the size of the affected area were among the reasons included in choosing IV over oral treatment. These clinical features made physicians were more likely, however, to choose inpatient over outpatient IV antibiotics.
  • 96% of physicians prefer children under 6 months to be treated as inpatients.
  • 46% of physicians prefer children aged 6 months to 1 year to be treated as inpatients.
  • Clinical features such as rash and fever deter 50% of physicians from using outpatient based IV treatment.
  • Fear of complications such as a deterioration going unnoticed, needing to re-present, or missing complications were reasons to prefer inpatient therapy 20-25% of the time.

Where next

  • Clinicians are receptive to home treatment with IV antibiotic therapy. Barriers to choosing home therapy should be addressed with education.

Additional resources

Summary by Emer Ryan, May 2020

Article

Biondi EA, Mischler M, Jerardi KE, et al. Blood Culture Time to Positivity in Febrile Infants With Bacteremia. JAMA Pediatr. 2014;168(9):844-849.doi:10.1001/jamapediatrics.2014.895 

How it helps

  • The febrile infant is a common presentation, however bacteraemia is uncommon in otherwise well infants. 
  • In febrile infants, most pathogens will be identified via blood culture within 24 hours, in previously well (non-ICU/non-surgical) patients. 
  • Duration of observation and empiric treatment vary widely amongst institutions. 

Want more…?

Study type/population

Multi-centre, retrospective cross-sectional evaluation of blood culture time to positivity in febrile infants (< 90 days age), with bacteraemia and no surgical/ICU history. 

Major findings

  • 5332 positive blood cultures were initially identified. 3129 were excluded from ICU and surgical patients and a further 1447 from those not treated for pathogenic organisms (common contaminants) were also excluded.
  • 392 blood cultures for which patient were treated with a full course of antibiotics were used in the analysis.
  • By 24 hours, 355 of the 392 blood culture results (91%; 95% CI, 88-93) were positive. By 36 & 48 hours, 378 (96%) and 386 (99%) were positive
  • E.coli was the most common organism identified (41%). Group B strep had the shortest median time to positive culture – 10.5 hours.

Where next

  • This study looked at 17 paediatric centres in the US. This may not translate to all other settings, including low resource settings. 
  • A study must include negative culture results and clinical outcomes (e.g. clinical deterioration) to comment on overall risk for serious bacterial infections in infants. 

Additional resources

Summary by Julian Dascalu, July 2020 

Article

Hoberman A, Paradise JL, Rockette HE, Kearney DH, Bhatnagar S, Shope TR, Martin JM, Kurs-Lasky M, Copelli SJ, Colborn DK, Block SL. Shortened antimicrobial treatment for acute otitis media in young children. New England Journal of Medicine. 2016 Dec 22;375(25):2446-56. 

How it helps

  • Reducing the duration of treatment for acute otitis media in children less than 2 years old with amoxicillin-clavulanate from 10 days to 5 days was more likely to result in treatment failure.
  • There was no change to the rate of adverse events or emergence of resistance.

Want more…?

Study type

A non-inferiority randomized, placebo-controlled trial across multiple sites (Children’s Hospital of Pittsburgh, paediatric practices and Kentucky Paediatric and Adult Research in Bardstown, Kentucky). 

Major findings

  • Acute Otitis Media (AOM) was diagnosed based on the AOM-Severity of Symptoms (AOM-SOS) scale, middle-ear effusion and appearance on otoscopy.
  • 520 children, 6 to 23 months of age with AOM received standard-dose amoxicillin-clavulanate for either 10 days, or for 5 days followed by placebo for 5 days.
  • Children who were treated for 5 days were more likely to have clinical treatment failure (77 of 229, 34% vs. 39 of 238, 16% - a difference of 17 percentage points between groups, 95% Cl of 9 to 25).
  • Clinical treatment failure included those with persistent and worsening clinical symptoms or signs at the end of treatment and was more likely with high initial AOM-SOS or recent antibiotics.
  • No significant between-group differences in rates of recurrence, adverse events or nasopharyngeal colonization with resistant pathogens.

Where next

  • Many guidelines such as the RCH CPG still recommend 5 days of amoxicillin if we are going to treat and no contraindications. Be aware some children will benefit from longer treatment 
  • This study does not tell us what we should do about older children, or in other low resource settings. 

Additional resources

Summary by Nelson Wang, July 2020

Article

Gurbani N, Dye TJ, Dougherty K, Jain S, Horn PS, Simakajornboon N. Improvement of parasomnias after treatment of restless leg syndrome/periodic limb movement disorder in children. Journal of Clinical Sleep Medicine. 2019 May 15;15(5):743-8.

How it helps

Children with restless leg syndrome (RLS) and periodic limb movement disorder (PLMD) also commonly experience parasomnias that occur during non-rapid eye movement (non-REM) sleep.

One of the treatments for RLS and PLMD is supplemental iron.

This study showed that treating RLS/PLMD with iron, also usually resolved non-REM parasomnias.  Suggesting that the two conditions exacerbate, and may even cause, one another.

Want more…?

Study type

This retrospective single site trial analysed one hospital’s records – to identify children with either confirmed RLS or PLMD who were followed up for at least 1 year.

50 of the 226 children identified also experienced parasomnias.

30 of these 50 children were treated with iron supplements (those with ferritin levels <50ng/mL). The other 20 were treated with other therapies.

The investigators analysed the sleep studies of all 30 children after their respective therapies - comparing the frequency of both parasomnias and RLS/PLMD events post treatments.

Major findings

Iron therapy reduced most children’s RLS symptoms, parasomnias and disrupted sleep.

Of the children taking iron supplements:

  • 50% had less RLS symptoms. 40% had no more parasomnias.
  • Most (2/3rds) noted these improvements within 3-6months of starting iron therapy.
    • Others took 6 – 24months.
  • But once achieved, the improvements stuck.
  • The post-therapy sleep studies also showed that these children had better sleep maintenance; fewer episodes of arousal (both the kinds that wake you up and those that happen whilst you’re still asleep - but disrupt sleep quality).

Where next

  • In general terms, parasomnias can improve over time (without intervention). The study does not include the results of the children who did not take iron supplements. Were their rates of improvement similar?
  • We know that the amount of iron the body needs to prevent anaemia, changes with age. Is the same true for children with either parasomnias or RLS/PLMD? Do children of different ages, sizes, or with different sleep-issues, require different amounts of iron?

Additional resources

Summary by Jye Gard, August 2021

Article

Kotagal S, Nichols CD, Grigg-Damberger MM, Marcus CL, Witmans MB, Kirk VG, D'Andrea LA, Hoban TF. Non-respiratory indications for polysomnography and related procedures in children: an evidence-based review. Sleep. 2012 Nov 1;35(11):1451-66.

How it helps

Polysomnography (PSG) is commonly used to investigate for obstructive sleep apnoea or other night-time breathing disorders.

However there are many other sleep disorders including hypersomnias (e.g. narcolepsy), parasomnias (e.g. night terrors), sleep-related movement disorders (e.g. periodic limb movement disorders), bruxism (i.e. teeth grinding) and enuresis.

This comprehensive literature review analyses the evidence for the role of PSG in these other sleep disorders.

Want more…?

Study type

A group of American Paediatric Sleep Physicians conducted a systematic review of the literature regarding non-respiratory sleep disorders and polysomnography (PSG). A single database (PubMed) retrieved 4,450 total English articles (published from 1966 to September 2010) from studies of children with at least 10 participants who underwent PSG.

The investigators found significant variability in the methodologies between studies. They used a grading system to weight the significance of different study results and guide their interpretation.

Major findings

The study provided evidence for the use of PSG to investigate

  1. Excessive daytime sleepiness or hypersomnolence in children >5 years old
  2. Unusual parasomnias. Such as those that occur more than twice weekly, are atypical (for ages, in onset, duration, frequency), or associated with movements that may be seizure or cause harm.
  3. Parasomnias, especially non-REM parasomnias such as sleep walking and night terrors, in children likely to have OSA or sleep related movement disorder
  4. Concerns that a child’s pain (e.g. due to juvenile arthritis) may be affecting their quality of sleep
  5. Often two or more PSGs are required to help investigate these four indications.

More about sleep disorders

Hypersomnolence or excessive daytime sleepiness is tricky to tease out in children, as the expected amount of sleep and napping changes with age, especially puberty (more here).

In general terms, sleep occurs in 90 - 120 minute cycles and two phases.

Every cycle of consists of:

  • Non-rapid eye movement (non-REM) sleep. The initial stages of sleep as you drift from being awake to progressively “deeper” sleep, and it becomes harder to wake you up.
  • Rapid eye movement (REM) sleep. When vivid dreaming and atonia occurs.

Parasomnias are abnormal sleep-related events such as movements, behaviours, emotions, dreams or autonomic activity that disrupts sleep, causes injuries or psychosocial stress.

  • Non-REM parasomnias include: Confusional arousals, sleep terrors, sleep walking
    They occur because the child becomes partially aroused from slow wave sleep –typically during the first half of the night the parasomnia will occur and then stops when a child returns to deep sleep. Common causes of non-REM parasomnias include anxiety, sleep deprivation, medicines, and obstructive sleep apnoea.
  • REM parasomnias include: Nightmare disorder, sleep paralysis

Sleep-related movement disorders, like periodic limb movement disorder, cause children to move about during sleep – disrupting the quality of sleep. Often they need to be distinguished from night time seizures.

Where next

  • Are there roles for PSG in children with other conditions (e.g. burns or headache syndromes that may wake them or disturb quality sleep)?
  • Are ambulatory sleep studies or other home monitoring devices just as efficacious?

Additional resources

Summary by Jye Gard, September 2021

Article

L'Huillier AG, Mardegan C, Cordey S, Luterbacher F, Papis S, Hugon F, Kaiser L, Gervaix A, Posfay-Barbe K, Galetto-Lacour A. Enterovirus, parechovirus, adenovirus and herpes virus type 6 viraemia in fever without source. Archives of Disease in Childhood. 2020 Feb 1;105(2):180-6.

How it helps

  • Routine point-of-care plasma viral PCR testing (a blood test that provides real-time results) showed that a third of children < 3 years old who present with fever of unknown origin (PUO) have one of four common viral infections (adenovirus, parechovirus, enterovirus or herpesvirus type 6).
  • A positive result helped clinicians discharge children home from the Emergency Department or keep their hospitalisations short. However, it didn’t stop clinicians from prescribing empirical antibiotics.

Want more…?

Study type

  • A prospective, two-year single-centre, epidemiological diagnostic study of <3 year old children who presented to the Emergency Rooms of Geneva University Hospitals, Switzerland, with fever without source (defined as T≥38 at home or in ER with no identified focus of infection after thorough history and clinical exam).
  • All children received a rapid-result plasma PCR test for four viruses.
  • These results were compared to a control group (50 children attending follow-up surgical outpatient appointments).

Major findings

  • In one third of children <3 years old with fever of unknown origin, at least one of either enterovirus, parechovirus, adenovirus or herpes virus type 6 was found on plasma viral PCR (34.8%, 47 out of 135 children)
  • The likelihood of having a serious bacterial infection was 5 times less if the child had a virus found on plasma viral PCR (only 2 of the 47 virus-positive children had a serious bacterial infection)
  • The likelihood of having a virus found on plasma PCR in a well child was low (3 children out of 50 controls)
  • Despite this, children with a positive plasma PCR result were equally as likely to be started on antibiotics as children who has negative results. However, the antibiotics prescribed were usually oral (rather than intravenous types).

Where next

  • A larger study might influence how we utilise viral PCR testing in investigating fever of unknown origin, including the prescription of empirical antibiotics
  • There is still a small risk of having viraemia and a serious bacterial infection (4.3%), so the detection of virus in the blood would not rule out the need for admission and antibiotics
  • But point-of-care plasma viral PCR testing may become part of our work up for fever of unknown origin and help stratify risk in <3 year olds who often present with non-specific symptoms
  • The point-of-care test used in this study examined viruses that are prevalent in the Northern hemisphere over summer months. What viruses should an Australian test screen for?

Summary by Sarah Bush, September 2020

Article

Morris R, Jones S, Banerjee S, Collinson A, Hagan H, Walsh H, Thornton G, Barnard I, Warren C, Reid J, Busfield A. Comparison of the management recommendations of the Kaiser Permanente neonatal early-onset sepsis risk calculator (SRC) with NICE guideline CG149 in infants≥ 34 weeks’ gestation who developed early-onset sepsis. Archives of Disease in Childhood-Fetal and Neonatal Edition. 2020 Mar 13.

How it helps

For infants >34 weeks gestation, the National Institute for Health and Care Excellence (NICE) guidelines and Kaiser Permanente neonatal Sepsis Risk Calculator (SRC) are good at detecting early onset neonatal sepsis. The NICE guidelines are superior at detecting the 1 in 5 cases that are asymptomatic at 4 hours of life. Neither NICE or SRC are good at detecting sepsis <4 hours of life. Hourly patient reviews are.

Want more…?

Study type

Retrospective analysis of five maternity centres in the UK – their clinical notes, blood and CSF cultures results. The researchers analysed the sensitivity of two different tools in detecting neonatal sepsis. Out of a birth population of 142,333 babies born >34 weeks, 72 were found to have positive cultures. Only 70 were included in the study, as records for 2 babies were incomplete or couldn’t be found. All 70 received antibiotics, however only 43/70 were given antibiotics early (within 4 hours of life). The remaining 27/40 culture-positive babies became symptomatic >4 hours of life and were started on antibiotics when these symptoms were seen. The researchers then applied both the NICE and SRC tools to the cases to see if they recommended starting antibiotics, and if so, when.

Major findings

  • The NICE guidelines picked up 39/43 babies that needed antibiotics vs the SRC that picked up 27/43. The extra 4 cases that were missed by both tools were started on antibiotics because of changes to other clinical observations.
  • The 12 cases caught by the NICE guidelines but not the SRC were clinically well babies whose mother had a fever or had received (adequate) intrapartum antibiotics. The SRC didn’t pick up any cases the NICE guidelines did not.
  • The researchers found that the trade off for prophylactically treating 12 asymptomatic infants with suspected early onset sepsis was the unnecessary treatment of approximately 11386-16852 infants. It also noted that reasons commonly given for not following the tools were only using parts of the tools (and ignoring others) and “clinical nervousness”.

Where next

  • The study recommended that a new trial should assess the efficacy of using the NICE guidelines first (given their increased sensitivity) and then the SRC for positive cases (given its more comprehensive suggestions for follow-up observations) – to see if this works better
  • Are there other factors to consider regarding the detection of? Like the rate of false-negative blood cultures due to inadequate volumes of blood collected in the culture bottle (contentious) or other test collection issues

Additional resources

  • Check out the Kaiser Permanente Neonatal Early-Onset Sepsis calculator here
  • Check out the NICE interactive flowchart here and full guideline here
  • Proof that it is safe and feasible to implement at tertiary Australian hospitals
Summary by Jye Gard, June 2020

Article

Rivas-Fernández M, Izquierdo AD, Cassanello P, Balaguer A. Do probiotics help babies with infantile colic?. Archives of disease in childhood. 2019 Sep 1;104(9):919-23. 

How it helps

Use of a specific probiotic culture called Lactobacillus reuteri strain Deutsche Sammlung von Microorganismen (LR DSM) 17938 may reduce crying and fussing times in babies that are predominantly breastfed.  

LR DSM 17938 is usually found in different body sites including the skin, breast tissue, gastrointestinal and urinary tracts. Other sources include dairy and meat products. It is also sold commercially. 

Want more…?

Study type

Literature review using seven different databases revealed 29 primary papers and 4 meta-analyses that examined the use of probiotics in children with colic. The review was limited to articles published in English or European languages. 

Major findings

Rigorous trials have yet to return a clear cause or treatment for colic. One leading hypothesis is that colic may be caused by dysbiosis (microbial imbalance).   

Positive finding were only found in infants who were  >50% breastfed. Infants who were mostly formula fed did not show any change in behaviour.  

In studies that showed positive findings, mostly breastfed babies who used LR DSM 17938 for 21 days halved their crying and/or fussing timesIn some studies this was a decrease in crying time of about 25 minutes per day. Importantly, other studies showed no benefits from LR DSM 17938 use.  

There have not been any reported adverse effects of LR DSM 17938 in healthy infants to date.  

Other strains of pro-biotics have not been well studied. 

Where next

There are currently 9 active randomised control trials also analysing the use of LR DSM 17938 in infantile colic.  

It is important that these studies: 

  • Clarify the recommended dose of LR DSM 17938 infants may trial – the dose used by different studies was highly variable, from 1x10^8 colony-forming units orally/day to 5x10^8 CFU/day to 1x10^10 CFU/day for 21 days.  
  • Review the cost effectiveness of using a daily probiotic, given that the natural course of infantile colic is self-limiting and benign. This study estimates that only 1 out of 8 infants with colic will gain benefit from using this probiotic (NNT = 8).  

Additional resources

Summary by Caitlin Richardson, August 2020

Article

Valkenborghs SR, Noetel M, Hillman CH, Nilsson M, Smith JJ, Ortega FB, Lubans DR. The Impact of Physical Activity on Brain Structure and Function in Youth: A Systematic Review. Paediatrics. 2019; 144(4):e20184032.

How it helps

This systematic review evaluates all the studies to-date that used MRI to evaluate the effect of physical activity on brain structure and cognitive function.

If increased physical activity is proven to improve cognitive outcomes, this may motivate change to widespread physical inactivity.

Want more…?

Study type

A systematic review of 9 studies of MRI findings in neurotypical school-aged children, pre and post physical activity interventions.

Studies were identified via a search of six electronic databases of English-language articles published between January 2002 and March 2019. Studies were eligible if they reported statistical analysis of MRI-based changes in brain function or structure following exercise interventions.

7 RCTs and 2 prospective observational studies were identified for inclusion and independently assessed by two reviewers. Interventions across the studies were heterogeneous, consisting of moderate-to-vigorous physical activity either twice a week or each school day for 20-120 minutes, ranging across a duration of 3-9 months.

Major findings

In short: The only consistent finding was that the parietal cortex appeared to function more maturely on fMRI following any physical activity intervention. It was difficult to compare results between studies as the intervention types, duration and way researchers interpreted similar MRI findings differed.

In long: Participants who partook in any physical activity intervention had consistently decreased activation of their parietal lobe regions on post-intervention fMRIs (performed whilst the child completed a cognitive tasks). For the small number of studies that also assessed improved fitness as an outcome, “fitter” children reliably showed improved white matter integrity.

However, there were other inconsistent findings across studies. MRI changes noted in all other cortical regions, as well as changes seen in functional network synchrony and white matter integrity, differed. The review authors hypothesised that the inconsistencies between studies was likely related to:

  1. The heterogeneity of the physical interventions implemented.
  2. Conflicting interpretations of MRI changes. Some studies interpreted increased activation of brain regions as an increased ability to perform tasks, whilst other studies interpreted decreased activation of the same regions as more efficient or mature task execution.
  3. Studies which used task-based fMRI did not report consistently on task performance between the intervention and control groups in support of fMRI findings.
  4. Risk of either selection bias (9 studies), reporting bias (7) or attrition bias (6).

Where next

  • Future research needs to measure cardiorespiratory and muscular fitness as part of the physical exercise intervention.
  • Studies should also measure changes at intervals – to see if there is a time exposure relationship between exercise and changes to the brain.
  • Longer prospective studies would be useful to determine if exercise prevents the degeneration of other critical brain structures, such as the hippocampus.

Additional resources

Learn more about interpreting MRI findings: Soares JM, Magalhaes R, Moreira PS, Sousa A, Ganz E, Sampaio A, Alves V, Marques P, Sousa N. A hitchhiker’s guide to functional magnetic resonance imaging. Front Neurosc. https://doi.org/10.3389.fnins.2016.00515

Summary by Elizabeth Cole May 2021

Article

Phelan PD, Robertson CF, Olinsky A. The Melbourne Asthma Study: 1964-1999. The Journal of Allergy and Clinical Immunology. 2002 Feb;109(2):189-94 

How it helps

Children with mild wheezing episodes associated with respiratory tract infections (RTI) generally do not develop asthma in adolescence or adulthood and usually cease to have symptoms by early adolescence. Whereas, children with wheezing not associated with RTIs and those with severe asthma are more likely to have asthma symptoms that persist into adolescence and adulthood. 

Want more…?

Study type

Cohort study to assess the long-term outcomes of childhood wheezing in Melbourne school children randomly selected at 7-years-old during routine school medical examination. 

Participants were cohorted based on severity of asthma symptoms (none; <5 episodes associated with RTIs; >5 episodes associated with RTIs; wheezing without RTIs; severe symptoms) then followed up at 10, 14, 21, 28, 35 and 42 years of age. 

Major findings

  • Children in the milder groups, particularly those with wheezing episodes associated with RTIs, typically had resolution of symptoms by early adolescence and did not go on to develop asthma symptoms in adulthood.
  • The severity of childhood symptoms correlated to the severity of symptoms in adulthood. At 21yr of age, 95% of those in the most severe cohort continued to have symptoms.
  • The severe group showed reduced lung function (FEV1 and FEV1/FVC) by the age of 14, though this did not worsen over time. This early loss in function was measured at a time before anti-inflammatory agents were routinely used.
  • There was no difference in lung function between in the milder groups and the control group on long term follow up, leading the authors to postulate that children in these groups do not require therapy at preventing progressive disease or loss of lung function (eg. steroids).
  • Atopic disease in childhood increased the risk of more severe asthma in adulthood.

Where next

  • Further studies are needed to determine the impact oral prednisolone has on pre-school wheezers’ short, medium and long-term health outcomes
  • What is the role of prednisolone in treating children in the ‘sticky’ age range of 9 months to 2 years old?
  • Children who used steroids within 14 days of presentation were excluded – further studies regarding repeat or prolonged steroid therapy are needed

Additional resources

Summary by Ary Sudarmana, August 2020

Article

Franklin D, Shellshear D, Babl FE, Schlapbach LJ, Oakley E, Borland ML, Hoeppner T, George S, Craig S, Neutze J, Williams A. Multicentre, randomised trial to investigate early nasal high—flow therapy in paediatric acute hypoxaemic respiratory failure: a protocol for a randomised controlled trial—a Paediatric Acute Respiratory Intervention Study (PARIS 2). BMJ open. 2019 Dec 1; 9(12).

How it helps

Using high-flow reduces the need to transfer children with bronchiolitis to a PICU, but the number needed to treat (NNT) is 9.

Want more…?

Study type

Prospective, non-blinded, randomised controlled trial at multiple centres (Australian & New Zealand hospitals).

Major findings

  • High-flow is safe to use. Adverse events were low in both the low-flow and high-flow groups (both had 1 child each that developed a pneumothorax that didn't need draining). This means children on high-flow treatment don't necessarily need extra nursing cares.
  • Starting high-flow doesn't change how long a child stays in hospital (ward or ICU) or needs oxygen.
  • In bronchiolitis, high-flow improves a child's observations (heart rate, respiratory rate and oxygen saturations) and thus reduces MET calls.

Where next

  • The treatment failure rates were lower at hospitals without a PICU. Is this because these hospitals have a higher threshold to start high-flow? Is this appropriate?
  • 1/3 escalations from low-flow to high-flow treatment was made at a doctor's request, not because their observations had changed. What are the factors that make us choose to escalate care, even if a child's observations are the same? Are there other bedside observations we should measure or account for?

Additional resources

  • Check out the first PARIS trial here.

Summary by Jye Gard, April 2020

Article

O’Brien S, Craig S, Babl FE, Borland ML, Oakley E, Dalziel SR; Paediatric Research in Emergency Departments International Collaborative (PREDICT) Network, Rational use of high-flow therapy in infants with bronchiolitis. What do the latest trials tell us?’ A Paediatric Research in Emergency Departments International Collaborative perspective, J Paediatr Child Health. 2019 Jul; 55(7):746-752.

How it helps

Hypoxaemia (<91%) in bronchiolitis should be treated initially with low flow O2 via nasal prongs, up to 2L/min. High flow should be used as rescue therapy if no improvement over 4 hours on low flow. Early use of high flow O2 for work of breathing in the absence of hypoxaemia is not supported by evidence

Want more…?

Study type

Systematic review of the evidence base comparing immediate high flow O2 therapy to rescue high flow O2 therapy. Analysing systematic reviews and randomised controlled trials of HFNC therapy in infants with bronchiolitis from 1 January 2000 to 27 June 2018.

Major findings

  • Early use of high flow O2 did not reduce the rate of ICU admission, rate of intubation, duration of oxygen therapy or duration of stay.
  • Children who will benefit from high-flow, will do so within 4 hours. Their heart rate & respiratory rate will begin to normalise and their Paediatric Early Warning Score will reduce.
  • High flow is an effective rescue therapy for those failing to improve with low flow therapy with ~60% avoiding further escalation to CPAP and admission to ICU.
  • Initial use of high flow O2 did not provide a cost benefit.
  • High flow therapy is considered safe with no significant increase in adverse effects compared to low flow.

Gaps in the literature

  • Significant variability exists in the literature when comparing target saturations, criteria to escalate care, and flow rates and FiO2 for O2 therapy.
  • No studies exist comparing treatment of bronchiolitis with high flow, vs treatment of bronchiolitis with NO high flow.

Additional resources

Summary by Julian Dascalu, April 2020

Article

Riordan A. In children with respiratory symptoms are Mycoplasma pneumoniae PCR and serology clinically significant? Archives of disease in childhood. Education and practice edition. 2014 Aug;99(4):157 

How it helps

M. pneumoniae PCR and serology are performed to try and find the aetiology of respiratory illness.

Many children carry M. pneumoniae in the naso-pharynx asymptomatically. 

Neither positive mycoplasma PCR nor serum IgM are reliable indicators of clinically significant infection. 

Want more…?

Study type

Prospective cross-sectional observational study.
726 children aged 3 months to 16 years.

Nasopharyngeal aspirates were taken for M. pneumoniae PCR and blood for M. pneumoniae IgM 
Findings expressed using odds ratios (OR). 

Major findings

  PCR positive  PCR negative  Total 
Symptomatic  51   263  314 
Asymptomatic  85   281  366 

  • 21.2% of asymptomatic children and 16.2% of the symptomatic group were M. pneumoniae PCR positive.  
  • The OR of respiratory symptoms with PCR positivity was 0.64 (95% CI 0.44 to 9.9, p=0.11)  
  • No significant difference in PCR-positive results or IgM positive results (p=0.23) between asymptomatic and symptomatic groups 

Where next

  • Is this true for all pathogens on the commonly-used NPA viral PCR panel? 
  • Would it be more cost effective to cease testing for M. pneumoniae and instead treat all presumed bacterial pneumonia with macrolide antibiotics if there is no response to amoxicillin, or in severe disease, as per the British Thoracic Society Guidelines?  

Additional resources

Summary by Nelson Wang, July 2020 

Article

Foster SJ, Cooper MN, Oosterhof S, Borland ML. Oral prednisolone in preschool children with virus-associated wheeze: a prospective, randomised, double-blind, placebo-controlled trial. The Lancet Respiratory Medicine. 2018 Feb 1;6(2):97-106.

How it helps

Prednisolone reduces the time preschool aged children with viral-associated wheeze spend in-hospital to <7hours. Especially pre-schoolers who present with a severe wheeze, previous diagnosis of asthma or who use salbutamol at home. Personal and family history of atopy are not good indicators of prednisolone effect.

Want more…?

Study type

A non-inferiority randomised, double-blind, placebo-controlled trial at a single-site (Emergency Department of Princess Margaret Hospital in Perth, Australia). Testing that placebo was non-inferior to prednisolone.

After this was finished, the researchers then used the data to perform a post-hoc superiority study.

Major findings

  • The “time in-hospital” was calculated from when the child was given prednisolone given until the time they were ready for discharge. Prednisolone reduced the incidence of children who stayed >7hours in-hospital
  • Steroids did not reduce the incidence of short (<4hours) hospital stays
  • Other secondary health outcomes were reduced: hospital admission, emergency re-presentations, general practitioner consultations, and use of salbutamol within 7 of the initial presentation to the ED.
  • The study used children aged 24-72 months so as not to confuse pre-wheeze with bronchiolitis. This hasn’t been done in many other studies!
  • No serious steroid related side-effects were reported (one child had a rash and two became hyperactive – one of the two was from the placebo group)

Where next

  • Further studies are needed to determine the impact oral prednisolone has on pre-school wheezers’ short, medium and long-term health outcomes
  • What is the role of prednisolone in treating children in the ‘sticky’ age range of 9 months to 2 years old?
  • Children who used steroids within 14 days of presentation were excluded – further studies regarding repeat or prolonged steroid therapy are needed

Additional resources

Summary by Jye Gard, May 2020

Article

Florin TA, Ambroggio L, Brokamp C, Zhang Y, Rattan M, Crotty E, Belsky MA, Krueger S, Epperson TN, Kachelmeyer A, Ruddy R. Biomarkers and Disease Severity in Children With Community-Acquired Pneumonia. Pediatrics. 2020 Jun 1;145(6). 

How it helps

FEB, neutrophil counts, CRP and procalcitonin  on admission were similar between cases of mild, moderate and severe pneumonia.  CRP and procalcitonin levels may be helpful in determining which children with severe pneumonia will need extra help (IV fluids, positive pressure ventilation, broad spectrum antibiotics, vasopressors and/or ICU admission). 

Want more…?

Study type

Single site (Cincinnati Children’s Hospital Medical Center Emergency Department) prospective cohort study of 477 children aged 3 months to 18 years old who presented between July 2013 and December 2017.  In order to be included, the children has to have: 
  1. One or more symptoms or signs of a LRTI (new or different cough, sputum production, chest pain, dyspnoea, tachypnoea or abnormal findings on auscultation) AND 
  2. A chest X-ray that showed focal opacity AND 
  3. Blood investigations
Note that in the US, X-rays and blood tests are more commonly done in children who have signs or symptoms of a LRTI. In Australia, we advocate that for clear-cut clinical pneumonia, there is no need to do them!  The researchers excluded children who had already been to hospital in the last fortnight or had underlying conditions that increased their risk of pneumonia (i.e. a history of aspiration pneumonia, immunodeficiency or other chronic diseases like neuromuscular disorders)   Severity of pneumonia was determined by the type of hospital treatment received (eg mild – discharged, mild-moderated – admitted, moderate-severe – admitted plus other therapy eg IV fluid, oxygen and severe – ICU) 

Major findings

  • The inflammatory marker values were very similar between pneumonia of different severity.
  • CRP and procalcitonin however seemed to be slightly higher in children who developed an empyema that required a chest drain and/or sepsis requiring vasopressors.
  • Some things that made interpretation of results a little sticky:
    • The researchers analysed composite outcomes (i.e. “moderate-severe pneumonia” were lumped together - all cases where children either received an IV fluid bolus, >12 hours of continuous IV fluid support, oxygen, antibiotics broader than just penicillin presumed complications or sepsis, were looked at all together) 
    • The blood results analysed were those taken when they first presented to hospital. As we know, markers such as CRP change with time and this does not account for different length of illness at time of testing.  
  • As an aside, 1142 children had a chest X-ray even though they had signs and symptoms of pneumonia! 477/1142 had blood tests!   

Where next

  • This is a single study in a single context.
  • Pneumonia complications take time to develop. Are they better at predicting severity or treatment needs when a child has been unwell for a certain period of timeOr when looked at as a trend alongside clinical progress to anticipate treatment needs? 

Additional resources

  • Check out this podcast with A/Prof Ed Oakley about avoiding unnecessary tests 
Summary by Jye Gard, September 2020