Prehospital blood for children by Delphi consensus, and a biennial appraisal finding prehospital guidelines are getting better — with resuscitation compilations leading

Two new documents, both published 11 September, and both were invisible to yesterday’s sweep — the trailing-window rule established two days ago recovered them, as it did all four items in today’s literature brief. The index for 11 September has grown from 1,035 records to 3,805 in 48 hours.

The French-society sweep returned its first hit of the month — an Infectious Diseases Now paper on pneumococcal conjugate vaccines and carriage in Moroccan toddlers with acute otitis media, dated 12 September. It is original research, not guideline output, so the substantive position stands: SFAR, SPILF and SRLF have published no guidelines in thirteen days. The expanded society-acronym sweep — now including ERC, ILCOR, ISHLT and ELSO after yesterday’s scope audit — returned three hits, none of them guidelines.

1. Prehospital blood transfusion in children — expert consensus where the evidence is absent

Group · Published: Pediatric Prehospital Blood Transfusion Collaborative. Prehospital Emergency Care, 11 September 2026 — two days old Magill, Heintz, Finney, Nimmer, Antevy, Clukies, Guyette, Ignacio, Kothari, Shah, Siegler, Studnek, Su, Browne

Not a tracked society, but squarely inside primary scope — haemorrhagic shock and resuscitation — and it addresses a genuine void:

Hemorrhagic shock is a leading cause of early trauma-related mortality in children, yet implementation of prehospital blood transfusion in children is heterogenous and without consistent guidance. Limited pediatric-specific evidence and physiologic differences from adults have contributed to substantial variability among existing EMS protocols.

Modified Delphi, 20 experts across paediatric surgery, paediatric emergency medicine and EMS medicine. Consensus thresholds prespecified at ≥80% agreement for inclusion, <50% for exclusion. 208 candidate items evaluated, 35 recommendations accepted over six rounds of voting.

Eligibility for prehospital transfusion requires all three:

1. Known or suspected severe haemorrhage or major blood loss, and
2. At least one indicator of shock following first-line haemorrhage control, and
3. EMS clinical judgement of the need for transfusion

Shock indicators: physiologic compromise, haemodynamic instability, poor perfusion, low cardiac output, or an abnormal validated paediatric shock index.

Contraindications: do-not-resuscitate orders prohibiting transfusion, and non-haemorrhagic shock.

And an explicit failure to agree: “We were unable to reach consensus on routine inclusion of end-tidal carbon dioxide as an indication for transfusion.”

Interpretation. Three things here are worth more than the recommendations themselves.

“Poor perfusion” is listed as a shock indicator, and “low cardiac output” alongside it. This archive has spent a month tracking the same idea — most recently in today’s brief, where a dual frequentist-Bayesian meta-analysis put peripheral perfusion-guided resuscitation at a 97.2% posterior probability of reduced 28-day mortality in septic shock. A Delphi panel independently reaching for perfusion rather than a blood pressure threshold, in bleeding children, is the same instinct arriving from a different direction — and in a paediatric population it is the correct instinct, because blood pressure is the last thing to fall in a haemorrhaging child.

The failure to agree on ETCO₂ is the most informative line in the abstract, and it is unusually well-timed. Two days ago this archive sent an Anesthesiology cohort of 185,455 patients showing low intraoperative ETCO₂ independently associated with mortality after adjustment for both hypotension and minute ventilation — and I noted then that the authors claimed it as a marker for risk stratification, not a target. Here is a Delphi panel refusing to make it a trigger for an intervention. Those two positions are consistent, and both are right. ETCO₂ carries genuine prognostic information and is not thereby a treatment threshold. It is rare to see a consensus panel decline to operationalise a variable it clearly believes in, and the refusal is more credible than an 80% agreement would have been.

Listing “DNR orders prohibiting transfusion” as a contraindication is precise in a way that matters. Note the qualifier: not “DNR status” but an order that prohibits transfusion. That distinction is exactly what the last three days of this archive have been about — the Annals of Surgery code-status cohort where continuing a limitation tracked prognosis rather than causing death, and the RAPM fracture analgesia study where a DNR order was associated with a third the odds of a rib fracture block in the least frail patients — a comfort measure withheld on the strength of a resuscitation order. A prehospital guideline that says the order must actually prohibit the transfusion rather than the patient has an order is drafting against precisely that failure mode. Whether crews read it that carefully at 3 a.m. is another question.

The honest caveat: this is expert opinion, and the abstract says so. “Limited pediatric-specific evidence” is the stated reason the document exists. Twenty experts, no grading of evidence certainty mentioned, 35 recommendations from 208 candidates — and the conclusion asks for “future prospective validation studies.” The third eligibility criterion is EMS clinical judgement, which is both unavoidable and an admission that the first two criteria do not decide the case. Treat this as a floor for protocol-writing, not as evidence that prehospital transfusion in children improves outcomes; that trial has not been done.

Read the guidelines · PMID 42725789

2. Prehospital guidelines appraised — 71 documents, and the resuscitation compilations are the good ones

Group · Published: Prehospital Guidelines Consortium. Prehospital Emergency Care, 11 September 2026 — two days old Martin-Gill, Patterson, Richards, Bishop, Potts, Cash

A biennial systematic review of prehospital evidence-based guidelines, Ovid Medline, EMBASE and CINAHL, 1 January 2023 to 21 May 2025, appraised against adapted National Academy of Medicine criteria and AGREE II.

71 new evidence-based guidelines identified.

• 67 (94.4%) addressed clinical topics, with 45 (63.4%) on time-life critical conditions and 37 (52.1%) each on injury and special clinical considerations
• 37 (52.1%) met all adapted NAM criteria for high quality — and of those, 32 (86.5%) were compilations of guidelines on cardiopulmonary resuscitation, emergency cardiovascular care, and trauma
• Mean AGREE II overall domain score 70.9% (SD 11.4%); 30 (42.3%) scored above 75%
• Incremental improvement over the 178 guidelines appraised in previous reviews

Interpretation. Read this directly against the SENTINEL appraisal sent four days ago, because the two reach opposite conclusions about guideline quality and both are probably right.

The SENTINEL International study appraised 14 hospital-specific paediatric sepsis guidelines from 13 countries and found every one of low methodological quality, with rigour of development scoring 0–50% and editorial independence 0–58%. This review appraises 71 prehospital guidelines and finds 52% meeting all NAM high-quality criteria, a mean AGREE II score of 70.9%, and measurable improvement over time.

The reconciliation is in what each was looking at. SENTINEL examined local hospital protocols — documents written by a department, for a department, with no systematic search and no conflict-of-interest process, because nobody expects one of a ward protocol. This review examined published evidence-based guidelines from organised bodies. Those are different objects, and the comparison tells you something useful: the guideline enterprise is working reasonably well at the level where it is done professionally, and badly at the level where care is actually delivered. SENTINEL’s finding that local protocols nonetheless transmitted the settled recommendations correctly — 93% agreement on time to antibiotics — is the other half of that picture.

The finding I would act on is the concentration of quality. Of the 37 high-quality guidelines, 32 were compilations on CPR, emergency cardiovascular care and trauma. That is not a coincidence: resuscitation is the field with ILCOR, a standing international evidence- synthesis apparatus producing CoSTR documents on a rolling cycle, which is exactly the infrastructure AGREE II rewards. Yesterday’s ERC 2025 backfill described that architecture in detail — ILCOR synthesises, ERC and AHA each write from it. This appraisal is independent evidence that the architecture produces better documents than the alternative, and it is an argument for other specialties copying it rather than each society reviewing the literature afresh.

The caveats. A mean AGREE II of 70.9% with SD 11.4% means a substantial tail below 60%, and only 42% scored above 75% — so “higher quality than before” is a modest claim, not a good result. And there is an obvious structural point: the Prehospital Guidelines Consortium is appraising the output of the field it represents, and reports improvement. That is not a reason to dismiss the numbers — AGREE II is a published instrument and the scores are checkable — but it is a reason to note who is holding the ruler.

Read the review · PMID 42725780

---

Blocked: the *European Heart Journal – Cardiovascular Imaging piece on sex differences in bicuspid valvulo-aortopathy within the contemporary international classification consensus (Michelena and Evangelista, 12 September, DOI 10.1093/ehjci/jeag253, PMID 42728840) has no abstract deposited and appears to be editorial commentary. It would have been directly useful — BAV phenotype definitions are a named gap in the partially-covered EACTS/STS aortic organ guidelines on the outstanding list — so it has been added there rather than characterised from its title.*

Also seen and not written up: the German Heart Surgery Report 2025 from the German Society for Thoracic and Cardiovascular Surgery (Thorac Cardiovasc Surg, 11 September) — a national registry report rather than a guideline, though worth a look on a quiet day for its volume and outcome denominators; author corrections to the Canadian Cardiovascular Society 2022 guidelines on cardiovascular interventions in adult congenital heart disease (relevant to the STS-ACSD reoperation model sent 11 September, but a correction notice carries no content); and the AE guidelines 2026 on antimicrobial therapy of periprosthetic joint infection, published in both summary and full versions in German — orthopaedic infection, secondary at best for this scope. Out of scope: AANEM carpal tunnel grading, WHO infertility guidance, two PREFERENDO paediatric endoscopy consensus documents, a Spanish Arteriosclerosis Society consensus on atherogenic dyslipidaemia, a German Delphi on COMT inhibitors in Parkinson’s disease, and several oncology and urology consensus statements.