The clinical context
Children O and P were two of three triplet boys born at the Countess of Chester in June 2016. The third triplet, Child Q, was later the subject of an attempted-murder count on which the jury failed to reach a verdict. O and P both collapsed and died within 24 hours of each other. They were term infants in the sense that any triplet pregnancy is — i.e. delivered earlier than a singleton term, but not extremely preterm. Triplet pregnancies carry significantly elevated complication rates.
Both babies underwent extensive resuscitation efforts. Neonatal resuscitation is a physically vigorous intervention: chest compressions, positive-pressure ventilation, high-volume fluid boluses into small circulations, and in the most intense cases direct cardiac massage. Prolonged or repeated resuscitation of a term newborn can produce a specific pattern of injury that paediatric pathologists recognise.
What the prosecution said about the post-mortem findings
Dr Andreas Marnerides, the Crown’s pathology expert, interpreted the post-mortem liver findings on Child O — in particular a pattern of hepatic injury and internal bleeding — as consistent with deliberate blunt impact. The prosecution argued that this was evidence of deliberately inflicted trauma, associated with alleged air-embolism mechanisms on the other indicted triplets. The interpretation rested on the premise that the pattern could not be adequately explained by the resuscitation efforts alone.
For our summary of his testimony, see the Marnerides pathology summary.
What independent pathologists say about the same findings
Independent paediatric pathologists who have reviewed the post-mortem material for the Shoo Lee Panel read the same findings differently. Their position can be reduced to four points:
- Resuscitation injury is a recognised pattern. The paediatric-pathology literature describes specific liver injuries occurring during vigorous neonatal resuscitation, particularly in term babies. The pattern includes sub-capsular haematoma, parenchymal contusion, and capsular tears. These are not rare findings in babies who have had prolonged CPR.
- Triplet gestation increases susceptibility. Triplet pregnancies are associated with placental insufficiency, fetal growth restriction and elevated susceptibility to circulatory collapse. A triplet baby who has deteriorated enough to require vigorous resuscitation is exactly the population in which resuscitation injury is most likely to present.
- Differential diagnosis was not adequately canvassed at trial. The prosecution pathology evidence was presented as diagnostic of deliberate trauma. An independent paediatric-pathology review conducted blinded to the case would, on the Panel’s view, have included resuscitation injury as a primary differential.
- Anatomical localisation. The localisation of the liver findings in Child O is consistent with the pattern seen in resuscitation-associated injury, not the pattern typically associated with focal deliberate impact.
Why this matters for the conviction
The liver-trauma finding on Child O is one of the more specific forensic-pathology elements in the Crown’s case. Unlike the air-embolism skin signs or the insulin immunoassay, it is a physical finding at autopsy with a paper record. It is therefore reviewable, and independent pathologists have reviewed it.
If the Panel’s reading is correct — that the findings are explicable by the prolonged resuscitation efforts these babies received — then the pathology-based element of the conviction collapses. The Crown’s case on O and P also included air-embolism-pattern evidence and the shift chart, but the liver-trauma finding was the element that was materially closest to forensic-standard proof.
A wider pattern
The resuscitation-injury interpretation is not unique to O and P. Across multiple indicted cases, the Panel’s reading is that post-mortem findings presented at trial as diagnostic of deliberate harm are, under a differential-diagnosis framework, better explained by the natural sequence of serious neonatal deterioration and the interventions that were used to try to save the baby. The underlying methodological issue — interpreting non-specific findings as specific to a prosecution theory, without blinded differential diagnosis — recurs. See evidence: expert methodology at trial for the wider pattern.
What the CCRC will have to decide
The CCRC application filed by Mark McDonald KC in October 2025 includes independent paediatric-pathology reports on the post-mortem material. If the Commission is satisfied there is a real possibility that a properly directed Court of Appeal would not uphold convictions built on the disputed pathology interpretation, referral follows. The pathology strand is, in this sense, one of the cleanest evidentiary strands for the Commission to examine — because the underlying physical material is preserved and reviewable in a way that the insulin samples and the clinical skin-sign descriptions are not.