Shaken baby syndrome: evidence and experts

cause of infant death, and Dr Squier does quote literature in which shaking was observed to have been associated with infant death. Presumably, this literature negates the discussions within the biomechanical literature; regardless of the data produced by animal models and dolls, real life has demonstrated that shaking can produce a sufficient force to kill an infant. However, what is clear from the literature quoted is that observed fatal shaking of an infant is a rare occurrence. In this review, much of the discussion of shaking focuses around biomechanical analysis of animals and models. When discussing the data generated by Ommaya et al., Dr Squier rightly reminds the reader of the differences between a rhesus monkey and a human infant. Similar statements should be applied to the dolls used in the studies cited in the later discussions. Many biomechanical studies are still using diffuse traumatic axonal injury as a benchmark for analysis of force generation. As far as we are aware, we still have no indication of the forces required to produce focal cervicomedullary axonal injury, which may underpin the observed hypoxic–ischaemic brain injury and, as such, cannot provide definitive statements. Injuries to the soft tissues of the neck, as well as to the cervico-medullary junction and cervical spinal roots, are undoubtedly apparent in some infants who present with the triad and these injuries have been documented. They may be missed if they are not looked for in each case. It is clear from the work of Geddes et al.1 and others that knowledge regarding brain injury in the infant is far from complete and it cannot be assumed that an infant brain responds to injury in the same way as that of an adult. That trauma can cause torn bridging veins and, therefore, SDH is beyond doubt and has been demonstrated at autopsy in cases of alleged non-accidental injury.2,3 In neonatal SDH, torn bridging veins are rare but trauma is the cause of the majority, through dural tears involving sinuses. Vitamin K administration is a recognized non-traumatic cause of intracranial haemorrhage in neonates but we are unaware of any other non-traumatic causes seen in clinical practice in this group. Therefore, it is difficult to argue against trauma as a mechanism for SDH in infants. Is hypoxia another likely cause of macroscopically visible SDH? There are papers based on observation at autopsy which refute that hypothesis4 but none supporting. Dr Squier dismisses this paper although we are uncertain which aspects of the clinical details would make the observations either more robust or unfounded. Therefore, until there is a definite body of evidence in support, this must remain a hypothesis only. Indeed this hypothesis was considered by the UK Court of Appeal and found to be wanting.5 The papers presented in the review actually refer to microscopic intradural bleeding6 (a common occurrence) or to a mathematical-computer model.7 Shaken baby syndrome: evidence and experts om m etary