We recently carried out a retrospective evaluation of 17 child abuse cases in the Norwegian courts and concluded that 16 of those failed to support the evidence provided by expert witnesses of shaken baby syndrome.1 Our paper focussed on the possible mechanisms behind the children's intracranial injuries and unfortunately a couple of findings were missing or incompletely described. In case number 2, we described rib fractures with callus formation. This was true for some, but not all, of the fractures. We did not mention the term ‘flail chest’, which we have now found in the discharge record. This was not mentioned elsewhere on the medical chart, as a finding or medical condition. The same is true for the bilateral rather than unilateral retinal haemorrhages, which again only were described in the discharge records. In case number 6, we described clavicle and rib fractures, but unfortunately, we did not report that there was also a scapula fracture. Stray-Pedersen et al.2 have claimed that we made other errors. We can explain these by stating, as is also described in the article, that we did not have access to all the data at the time of the retrospective analysis. For example, we made a deliberate decision not to view any legal documents, in order to avoid introducing bias. These discrepancies included the fact that we did not mention the bilateral optic nerve sheath haematomas or metaphyseal fractures in case 4 because the medical charts did not contain this information. Furthermore, Stray-Pedersen et al. have stated that there was a rib fracture in case 4. This was suspected during the autopsy but could not be verified by microscopical examination. With regard to case 6, Stray-Pedersen et al stated that ‘the defendant confessed to having repeatedly shaken the child back and forth’. This information was not given in the article, as it was only mentioned in the legal documents and not in the medical records. However, following a reconstruction in the court, two medical experts, including Stray-Pedersen, concluded that the shaking that was demonstrated had insufficient power to cause intracranial findings. In addition, Stray-Pedersen et al. have disagreed with us on a few other cases, where we stand by our interpretations. For example, in case two, they claim that we have overlooked fresh haemorrhagic components in the subdural effusions. We have simply called these chronic subdural hematomas/hygromas, which we believe is reasonable since small rebledings in chronic SDHs are a common finding. Further, in case 4, which involved a one-month-old infant, Stray-Pedersen et al. claim that there was an acute subdural hematoma. We were unable to detect this on the images, it was not described in the medical charts, and we still find it hard to detect this finding. Our assessment of the low-quality computed tomography images for case 6 was that no definite acute, bilateral subdural haematoma was present, contrary to the view expressed by Stray-Pedersen et al. They also claim that the fractures described in case 6 were new. We felt that this was not the case, because callus was observed after another 11 days. In conclusion, we regret any inadvertent errors in our paper. The aim of our study was to analyse the intracranial findings presented by the court experts involved in these 17 cases and to evaluate if they showed evidence of traumatic origin per se. However, while it is important to present data as accurately and objectively as possible, it is also important to acknowledge the difficulties that arise in retrospective studies. We must stress that we are fully aware of the fact that child abuse exists, and of the importance of protecting abused infants. It is further not our conclusion that abuse can be excluded in all cases in this cohort. In the evaluation of these cases, it is self-evident that extracranial injuries such as bruises and different fractures must be taken into consideration. A comprehensive assessment of these was however beyond the scope of this article. That being said, we believe that the findings of our study still hold and that many of the intracranial findings in this cohort of possibly abused infants could be attributed to non-traumatic causes. One important example quoted in our paper was the group with benign external hydrocephalus. While Stray-Pedersen et al have made some valid points about our paper, it is interesting to note that they did not question our main intracranial findings.
Read full abstract