This Week in Pediatrics — Aug 31, 2026

Welcome to This Week in Pediatrics. This week we're covering 10 notable papers spanning respiratory care from the delivery room to the operating theatre, risk stratification tools in sepsis, congenital infection and preterm follow-up, and a set of practical management questions from foreign body ingestion to vaccinating children with metabolic disease. Let's dive in.

We'll start with oxygen and the lung, because two papers this week bracket the whole span of paediatric respiratory support. In Pediatric Research, Gupta and colleagues report what they describe as the first randomised trial in children to test how much oxygen we should be giving just before we pull the tube. Sixty-four healthy children aged one to fourteen, having elective non-thoracic, non-abdominal surgery in the supine position, were randomised to either fifty percent or one hundred percent inspired oxygen before extubation, with lung ultrasound before and after. The result is striking in its cleanliness: not a single child in the fifty percent group developed atelectasis, compared with just over half of the children given pure oxygen. Nobody desaturated in either arm, and postoperative fever rates were the same. So even ten to twenty minutes of one hundred percent oxygen at emergence is enough to produce absorption atelectasis in a child's lung. The caveats matter — this is a single-centre trial in sixty-four healthy children, ultrasound atelectasis is a surrogate rather than a clinical outcome, and we don't know whether it translates into anything the child feels. But it directly challenges the reflex of dialling up to one hundred percent for emergence, at least in low-risk elective cases.

At the other end of life, the European Journal of Pediatrics carries a review by Courtney on non-invasive ventilation in the delivery room, tracing how we moved from intubating almost every baby in distress to starting with nasal CPAP as the first-line intervention, with non-invasive intermittent positive pressure ventilation used either when CPAP fails or from the outset. The honest message is how much remains unsettled — the ideal CPAP pressure, whether intermittent positive pressure ventilation is genuinely better, how to synchronise with an infant's own effort, and what any of this means for the extremely low birthweight baby. Read alongside the extubation trial, the theme is the same: less aggressive is increasingly the default, but the dose and the details are still being worked out. [1] [2]

The second theme is prediction — three papers this week try to turn messy clinical data into a number you can act on. In Pediatric Research, Velez and colleagues built a machine-learning tool for a very specific decision point in paediatric emergency care: the child screened for sepsis who has already had two fluid boluses, and the question of whether to keep giving fluid or escalate to vasoactives. Across five emergency departments and a final cohort of three hundred and forty-one children, about one in four went on to receive vasopressors. A random forest model using eight routinely available variables — with mean arterial pressure after the second bolus and baseline pressure severity doing most of the work, and blood urea nitrogen the only laboratory value retained — discriminated reasonably well and sorted children into four risk tiers, from under seven percent needing vasopressors in the lowest tier to nearly two thirds in the highest. This is retrospective and derivation-stage only; it has not been prospectively validated or shown to change outcomes. But it maps neatly onto a decision where guidelines are genuinely vague. [4]

Alongside that, in the European Journal of Pediatrics, Villaverde and colleagues present the cCMVnet score for congenital cytomegalovirus, drawn from a European registry of five hundred and seventy children across eleven countries. Roughly two thirds of those children were symptomatic at birth, and just under thirty percent developed sequelae — predominantly sensorineural hearing loss, then motor impairment, epilepsy and visual impairment. Splenomegaly, the severity of hearing loss, and neuroimaging abnormalities including white matter change, calcifications, ventriculomegaly and cortical malformations were the significant predictors, and the resulting score discriminated well with a negative predictive value around ninety percent at the proposed cut-off. The clinical appeal is that it stratifies beyond the crude symptomatic-versus-asymptomatic dichotomy, integrating audiology and imaging. It is internally validated only, and the authors are explicit that external validation is needed before you use it to counsel families. [3]

The third prediction paper, in the Journal of Pediatrics, comes from Derbie and colleagues, who built a perinatal social determinants of health composite score in three hundred and twenty-eight infants born at or before thirty-two weeks across five Cincinnati neonatal units. Five prenatal variables — maternal age, education, household income, employment and family structure — were combined by machine learning and tested against Bayley scores at two years. Infants in the highest social risk tertile scored about thirteen points lower on cognition and nineteen points lower on language than those in the lowest tertile, gradients that persisted on temporal validation against general cognitive ability at age five. Discrimination for frank impairment was only moderate, so this is not a diagnostic test — but as a tool for triaging early intervention and for case-mix adjustment when comparing neonatal units, the social gradient here is as large as many of the biological predictors we routinely chase. [9]

Our third theme is follow-up and complications, and Acta Paediatrica contributes two very different papers. De Gouveia Belinelo and colleagues followed ninety-nine very preterm infants for the first year after neonatal unit discharge with monthly symptom diaries and nasal swabs. Every single infant had at least one acute respiratory illness. Rhinovirus was the commonest organism overall, detected in over forty percent of illness swabs, but respiratory syncytial virus dominated the presentations that actually landed children in the emergency department or in hospital. Boys fared worse — roughly four times the hospitalisation risk and substantially more asthma medication use — and discharge during summer was associated with more illness and more admissions, a reminder that the timing of going home shapes the exposure window. The practical message is that post-discharge respiratory morbidity in this group is close to universal, and preventive strategies need to reflect that.

The other Acta Paediatrica paper, from Salatto and colleagues, is an Italian multicentre review of twenty years of surgical complications from swallowed foreign bodies — forty-two children, which they describe as the largest such series reported. Nearly three quarters had intestinal perforation, the remainder obstruction or prolonged retention requiring removal. What should stick with you is what was swallowed: magnets accounted for more than forty percent of cases, disc batteries about a fifth, and sharp objects another fifth. Magnets and button batteries are the objects that turn an observation decision into a surgical one. [8] [5]

Three remaining papers round out the week on management questions. In the European Journal of Pediatrics, Wang and colleagues report on three hundred and ninety-eight children with necrotising pneumonia, of whom about one in ten needed surgery after medical therapy failed. Mycoplasma pneumoniae was the leading pathogen, identified in just over half of cases — which is an argument for routine atypical pathogen testing. Over a median follow-up of more than five years there was no recurrence and no mortality, and the five postoperative pulmonary events all resolved with conservative management, supporting timely resection when medicine fails. In Developmental Medicine and Child Neurology, Barkey and colleagues report five-year outcomes after deep brain stimulation in children with dystonia and other hyperkinetic disorders. Self-selected goal performance and satisfaction on the Canadian Occupational Performance Measure improved substantially and were sustained, while the change in dystonia severity scores was comparatively modest — median motor scores fell from eighty-one to seventy, with most of that in the first year. Gains were largest in monogenic dystonia and in children with better baseline hand and gross motor function. The important nuance for counselling families, particularly in dystonic cerebral palsy, is that meaningful functional gains occurred despite limited reduction in measured dystonia. Finally, Lipiński and colleagues review vaccination in children with inborn errors of metabolism, where immunisation is often delayed for fear that post-vaccine fever or poor intake will trigger decompensation. The paediatric evidence is limited and mostly small observational work, but it shows no consistent safety signal in clinically stable children, and guidance supports vaccinating on the national schedule unless there is a vaccine-specific contraindication, with individualised decisions around live vaccines and serology. [6] [7] [10]

If you only have time for one paper this week, make it the Pediatric Research trial on inspired oxygen before extubation. It is a randomised, immediately actionable challenge to an almost universal anaesthetic habit, and the effect size is hard to ignore.

Here are the key takeaways from this week in Pediatrics. First, one hundred percent oxygen before extubation produced ultrasound atelectasis in more than half of healthy children, while fifty percent produced none, with no desaturations — reason to reconsider routine pre-extubation hyperoxia in low-risk cases. Second, two new risk scores, one for fluid-refractory septic shock and one for congenital cytomegalovirus sequelae, both perform respectably but are derivation-stage tools that need prospective and external validation before they drive decisions. Third, social disadvantage measured at birth predicts a thirteen to nineteen point spread in two-year Bayley scores among very preterm infants — worth capturing systematically in follow-up clinic. Fourth, essentially all very preterm infants get a respiratory illness in the first year home, with respiratory syncytial virus driving the admissions. And fifth, magnets and disc batteries dominate the ingestions that end in surgery, while vaccinating metabolically stable children with inborn errors of metabolism on schedule appears reassuringly safe.

That's your roundup for This Week in Pediatrics. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And if it saves you time, a quick rating in your podcast app helps other physicians find the show.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.