Subdural haemorrhage in a child under 2 years

Subdural haemorrhage in a child under 2 years

Children under 2yrs presenting with subdural haemorrhage (SDH) are an important cause of morbidity and mortality. A significant number will have been caused by purposeful, inflicted, trauma, as part of an acceleration/ deceleration injury. In the investigation of non-accidental head injury (NAHI) it is important to differentiate inflicted injury and other causes of SDH. This section aims to help doctors and other staff thoroughly investi-gate the child presenting with SDH.

·  This section assumes that the child is stable clinically (airway, breathing, circulation) and relevant teams are being contacted for further opinions.

·  The investigations will take at least a week to perform, therefore there is no ‘hurry’ to produce definitive report/guidance for other professionals until results are available.

·  Do not use the term ‘shaken baby syndrome’. Shaking is a possible mechanism of injury, and not a syndrome, and should be considered in the context of other mechanisms of non-accidental head injury.

·  NAHI is a leading cause of death and disability in children, particularly if cerebral injury is part of the spectrum of damage. Bleeding from torn bridging veins into the subdural space is the hallmark of non-accidental head injury. When infants who developed a SDH after infection or neurosurgical intervention are excluded, in retrospective studies, 24–82% of cases with SDH were highly suggestive of abuse in different series.

Differential diagnosis of SDH

·  Trauma, traumatic labour.

·  Neurosurgical complications, cranial malformation (aneurysm, arachnoid cyst).

·  Cerebral infections.

·  Coagulation and haematological disorders.

·  Metabolic (glutaric aciduria, galactosaemia).

·  Biochemical disorders (hypernatraemia).

Symptoms/signs of acute SDH

·  Encephalopathy (irritability, crying, inconsolability, unsettled behaviour, lethargy, meningism, decreased or increased tone, seizures, impaired consciousness).

·  Vomiting, poor feeding.

·  Breathing abnormalities, apnoea.

·  Pallor, shock.

·  Tense fontanelle.

Early post-traumatic seizures occur more frequently in inflicted than in non-inflicted head injury.

Symptoms/signs of subacute or chronic SDH

·Expanding head circumference.

·Vomiting, failure to thrive.

·Neurological deficit/s.

Retinal haemorrhages (‘haemorrhagic retinopathy’)

·Although strongly associated with NAHI, retinal haemorrhages are not specific for the diagnosis, nor can they be dated with precision. In NAHI haemorrhagic retinopathy can typically affect all retinal layers. It shows different ages and stages of resorption. It can be found throughout the retina to the ora serrata.

·Vitreous haemorrhage is frequent.

·Retinal haemorrhage may be unilateral or asymmetric in terms of number and distribution, however, some victims have none at all (15–25%). In severe life threatening trauma (motor cycle, great height) retinal haemorrhage is found in less then 3%. Retinal haemorrhages in newborns are seen in vacuum-assisted deliveries in up to 75% and in spontaneous vaginal deliveries in up to 33%. They resolve by 2wks after birth, at the latest by 6wks, in the great majority. A consultant ophthalmologist with expertise in the assessment of the eyes in children suspected to have NAHI should examine the child.

Fractures

·Since skull fractures may be missed by bony windows on CT, a plain skull film should be obtained.

·Skull fractures do not heal by callus formation and so dating of an injury is especially difficult. If the edges are round and smooth it is likely to be more than 2wks old.

·If the skull fracture is depressed or has branching, crossing, or stellate fracture lines, it is highly suggestive for NAI, whereas accidental fractures typically are linear, parietal, and over the vertex.

·The typical non-skull fracture of child abuse is the metaphyseal fracture caused by twisting the limb. It can also occur from birth injury (e.g. breech extraction). The ‘bucket handle’ and ‘corner’ type metaphyseal fractures are very suggestive for NAI. It is very important to target X-ray imaging on the metaphyses, as wider imaging can miss fractures.

Brain imaging

·The initial investigation is likely to be CT, but MRI will also be necessary in most cases (Fig. 14.4).

·MRI is more sensitive in identifying SDH’s of different signal characteristics, posterior and middle cranial fossa bleeds, and parenchymal changes in the brain.

CT scans may miss small subdural bleeds. Blood along the tentorium, interhemispheric haemorrhages, and SDHs in multiple sites or of different densities were almost exclusively seen in NAI. While acute haemorrhage may be isointense with brain on T1-weighted images, acute or subacute blood is more likely to be moderately hyperintense on these sequences.

·  T2-weighted sequences may also show high intensity, although this may be difficult to separate from the adjacent signal in CSF. The FLAIR sequence suppresses the signal from normal CSF, allowing the high signal haemorrhage to be visualized.

·  In the early acute stage when the blood clot is solid it may not be impressive. However, as it breaks down by fibrinolysis and water is drawn into the haematoma a marked effusion may become visible. Due to the dynamic changes of pathology sequential brain imaging is recommended in order to capture the evolution of different lesions. Within 2–4wks contusions and tears are at their most prominent.

·  Encephalomalacia may be apparent and even early atrophy. By 2–3mths atrophy is well established. Areas of contusion and hypoxia-ischaemia have evolved into cysts, and SDH should be clearing.

Coagulation and haematological disorders

·  For the exclusion of thrombocytopenia, anaemia and malignancy: do platelet count, FBC and blood film. Renal and liver function tests rule out these acquired coagulation defects.

·  The ‘coagulation screen’ comprises of PT, APPT, Thrombin time, Fibrinogen and ‘Mixing studies’ (50:50 mix) to exclude inhibitor.

Factor assays are available for factor II, V, VII, VIII, IX, XI, and ‘von Willebrand`s disease’. An α 2 antiplasmin deficiency is diagnosed by a thromboelastogram (TEG). Platelet function disorders, vitamin C

deficiency, Factor XIII, and collagen disorders are extremely difficult to diagnose in a child under 2yrs and are very rare conditions.

·Therefore, investigations for these disorders should only be done after discussion with Paediatric haematologist, and on good clinical grounds.

Glutaric aciduria type 1 (GA1)

The exclusion of GA1 is fraught with difficulty. The best approach is to obtain the urine and blood, however, to delay further investigation until other investigations are back. As an example, if the child has multiple frac-tures, or malicious injuries these would not have been caused by GA1. In such an instance the investigations for organic acids, acylcarnitines, or even a skin biopsy are inappropriate.

Management

·Take full social, medical, family history, including report from social services and police on all adults in household/caring for child.

·CT head and MRI head/spine when possible.

·Skeletal survey.

·Clotting assessment.

·Take urine to store in case of need to check organic acids.

·Arrange ophthalmology assessment.

Unless sure that there has been accidental trauma, arrange a full confer-ence around the child with relevant professionals, including social services, who may invite police attendance (their responsibility). Remind them: the investigations will take at least a week to perform, therefore there is no ‘hurry’ to produce definitive report/guidance for other professionals until results are available.