Abusive head trauma mechanisms myths and mysteries Jo

Abusive head trauma – mechanisms, myths and mysteries Jo Tully VFPMS seminar 2017 With thanks to Dr Ciara Earley, Dr Trusha Brys and Dr Anne Smith

What this talk is NOT…. . Short falls Direct impact trauma • Single event • Crush injuries • Can kill (but very rarely!!!) • Direct BFT to head • Almost always result in minimal to no injury • Very rarely cause intracranial injury • Sometimes result in scalp injury, skull fracture • single linear parietal bone • Focal SDH • Focal brain contusion • Lucid interval

What is “shaken baby syndrome? ” Whiplash shake syndrome Inflicted head injury Triad of; • SDH • RH • Encephalopathy Abusive head trauma Shakingimpact syndrome Shaken baby syndrome Nonaccidental head injury

HISTORY 1860 Ambroise Tardieu French physician Medical, psychiatric, social and demographic features of child abuse as a syndrome 1946 Dr Caffey Paediatrician/ Radiologist SDH and long bone fractures inconsistent with accidental cause (also noted retinal haemorrhages 1953 Silverman Multiple fractures from intentional trauma not organic disease 1962 Dr Kempe Paediatrician ‘Battered child syndrome 1971 Dr Norman Guthkelch Paed neurosurgeon UK Shaking as mechanism 1972 - 1974 Caffey Parent infant traumatic stress’ ‘Infant shaken whiplash syndrome 2009 American Academy of Paediatrics ‘Abusive head trauma’

“Pediatric abusive head trauma leading cause of child abuse deaths” “Battlelines drawn as shaken baby syndrome controversy set to run” “Dangerous Vaccines Found to Cause Symptoms of Shaken Baby Syndrome” “Shaken baby injuries rose in recession” New doubts are cast on shaken-baby diagnoses Accepted science is faulty, lawyers say

Clinical features and the ‘overlap’ • • Irritability Poor feeding Vomiting Lethargy Resp. comp Seizures Apnoea Delayed development • • Scalp injury Skull # Increasing HC ICH DAI Cerebral oedema C-spine fracture Cervical spinal cord injury/haemorrhage • Rib/long bone fractures • Retinal haemorrhages

Epidemiology and missed cases • 3 -4 severe/fatal cases per day in US (2004) • Homicide leading cause of trauma-related death <4’s • > males (56%) • Infants • Lower SE groups • Sentinel injuries/previous presentation • Jenny et al 1999 Recent systemic review Maguire et. al APNOEA important discriminator < 1 in 5 cases recognised if; • Normal respiration • No seizures • No skin injuries • Caucasian family with 2 parents • 30% cases “missed” • 30 -60% have # • Absence of bruising

The challenges facing clinicians • Majority of cases are young and non verbal • Family/clinicians unwilling to consider abuse • Are the findings a result of trauma or a medical condition? • Medical conditions can be confused with abuse • Always consider alternative possibilities • With AHT this is a VERY important aspect of care • If trauma, ? abuse, neglect or accidental • If trauma, What type, size, direction of mechanical force? • Controversies over cause/entity

The approach……. . • Detailed history surrounding • Presenting complaint • Trauma history • Psychosocial history • Physical examination • Other causes for presentation can be eliminated through detailed Hx, PE, initial lab/radiological imaging

Important anatomy Meninges – subdural space Bridging veins • travel up from surface of the cerebrum • pass through subdural space to sagittal sinus • Sagittal sinus fixed to inside surface of the skull 10

AHT – mechanisms • Acceleration deceleration / rotational forces • Impact – direct forces • Combination of impact and accelerationdeceleration Forces act on… • • Scalp Skull Spaces / CSF Meninges Brain Ventricles Brainstem / tentorium • Spine

What does shaking do to infants? Quietens them. . . “that worked” Repeated events (not a severe shake first up? ) Clinical spectrum; • Acute encephalopathy - severe shake – catastrophic event, sudden collapse, apnoea, arrest = VERY close in time to collapse • Sub-acute encephalopathy – increasing brain swelling - NOT 100% OK after the shake (symptoms might be subtle) • Sub-acute non-encephalopathy – fractures, bruising • Chronic – increasing HC & raised ICP

Subdural haemorrhages Birth related • Common • All modes > vaginal - ? oxytocin • Resolve by 4 weeks • Located posteriorly • Asymptomatic Inflicted • Thin films, multiple (OR 6) • Interhemispheric, parafalcine (OR 9. 5) • Over convexities (OR 4. 9) • Along tentorium • In posterior fossa (OR 2. 5) • Associated with HIE (OR 3. 7) and oedema Accidental • Localised, unilateral • May cause mass effect • Less likely to cause loss of consciousness, focal parenchymal injury • Extradural > accidental Neuroimaging; what neuroradiological features distinguish abusive from non-abusive head trauma? Kemp 2011

Other causes of SDH • • Accidental trauma • • Bleeding • diathesis/coagulopathy • • Medical conditions • Thorough history Examination Location/distribution Appropriate lab tests Radiology • Glutaric Aciduria Type 1 • Scurvy • Should exclude • Menkes kinky hair • Congenital malformations • Infectious : meningitis alternative causes – diagnosis of exclusion • What else could have produced these findings?

Retinal haemorrhages Ora serrata Cardinal feature of AHT, in 85% 15

Other causes of RH • Hypertension • • Bleeding disorder • • Meningitis/sepsis/end • ocarditis • • Leukaemia • Vasculitis • • Cerebral aneurysm • • Retinal disease CO poisoning Anaemia Raised ICP Glutaric aciduria, galactosaemia ECMO Hypoxia So how can we say these retinal haemorrhages were caused by abuse?

Pattern, type, extent and distribution…… • Multilayered • Pre-retinal, intra-retinal, sub-retinal • Extending to oro serrata • Too numerous to count • Flame, dot/blot • May have retinal folds/retinoschisis (splitting of retina) – increased in fatal cases

Multilayered RH

Forming an opinion on RH’s • Multilayered, numerous (AHT, severe MVA/crush injury, falls > 10 m) • Uncommon in accidents – unilateral, posterior pole, few in number • Do not occur with coughing, vomiting, seizures • Multilayered rarely persist beyond a few days, dating difficult • Common after birth (esp instrumentation) – majority resolve by 10 days (up to 2 months). If early flame/blot DD may be tricky. • Usually easily exclude other causes (cannot be confused) • Other causes (raised ICP, hypertension, resuscitation) • Scattered around posterior pole, intraretinal • Not same distribution as AHT Ophthalmic concerns in abusive head trauma J Fam Viol Levin 2016

The controversies surrounding AHT • The triad cannot be caused by shaking alone • Forces required to cause injury would damage neck • Legal perspective - If an infant is shaken what injuries would occur IN THIS case? • Geddes “unifying hypothesis” pathogenesis of SDH +RH was hypoxia ischaemia not trauma • Squires • Duhaime and early biomechanical studies Geddes J. F. et al Dural haemorrhage in non-traumatic infant deaths: does it explain bleeding in “shaken baby syndrome”? Neuropathol Appl Neurobiol 2003, 29: 14 -22 Squier W Shaken baby syndrome; the quest for evidence Dev Med Child Neurol Jan 2008; 50: 10 -14

The saga continues…. • Geddes - UK court of appeal 2005 – “the unified hypothesis can no longer be regarded as a credible or alternative cause of the triad of injuries” • Squires – struck off medical register in 2016 • Not yet got a credible alteranative hypothesis Squier W, Adams L. B. The triad of retinal haemorrhage, subdural haemorrhage and encephalopathy in an infant associated with evidence of physical injury is not the result of shaking, but is most likely to have been caused by a natural disease J. Prim Health Care 2011: 3(2)159 -163

ALTERNATIVE THEORIES • • • Birth related trauma rebleeds Hypoxia causes SDH (Geddes/Barnes) Sinus venous thrombosis Resuscitated SIDS leads to hypoxia leads to SDH Immunisations Benign enlargement of subarachnoid spaces of infancy (BESS) leads to signs confused with AHT • Retinal haemorrhages are not indicator AHT (vomiting, coughing) • Short falls (Plunkett) produce findings • Biomechanical studies – shaking alone does not generate forces required (Bandek, Durhaime) Jenny C Alternative theories of causation in abusive head trauma: What the science tells us Pediatr Radiol (2014) 44 (Suppl 4) S 543 -S 547

Birth related trauma and rebleeds • • Mixed density SDH’s Birth haemorrhages common Theoretically might re-bleed Re-bleeds generally: • No change/gradual change clinical status – not clinically catastrophic • Small • Usually infants with brain atrophy • From area of previous damage/hhge (not new area) • Usually asymptomatic

Hypoxia • Geddes et al. (theory) • SDH caused by hypoxia • RH caused by hypoxia (retinal exam not doc. in non traumatic grp) • Limited clinical data, did not know how specimens were collected - ? artefact • Confirmed hypoxia – imaged – SDH not seen • Barnes – infant sudden collapse/died – choked on formulae BUT authors failed to disclose #, successful conviction for AHT….

Sinus venous thrombosis • SDH attributed to primary SVT • Rare – adults with severe underlying illness • Newborns more susceptible to SVT and coexisting SDH often found • After postnatal period SDH from SVT (in absence trauma) not been reported • Die from severely raised ICP - SVT if ICP caused by trauma – causation cannot be indicated

Immunisations • Anti-vaccine groups – brain swelling/ICH dx as ‘shaken baby’ actually d/t vaccines • Multiple times in the 1 st year • AHT can occur after immunisation • Vasovagal – fall – some SDH • No evidence in literature vaccines cause any condition that can be confused with AHT

Signs AHT Benign enlargement of subarachnoid spaces (BESS) • Common and BENIGN • Infants unusually large heads • Immature balance between CSF prod/absorp expansion CSF within subarachnoid space • HC >50 th at birth, resolves 1 -2 years • Issues: • Can be confused with a SD hygroma • ? At increased risk of bleeding - neuro collapse, SDH and RH • May predispose to SDH after insig trauma? • Cases – SDH localised, asymptomatic, no RH • Isolated SDH in infant with BESS with no Sx and no associated stigmata of abuse may not indicate NAI

Other Subdural collections • Subdural hygroma vs chronic subdural haemorrhage • Radiological appearance similar to CSF • What is the pathogenesis? Are they all caused by trauma? • Trauma and resolution of SDH • Glutaric aciduria type 1 • Rupture of pre-existing arachnoid cyst – none described in infants • BESS • All infants with SD collection need AHT evaluation

Confessional data • 100% reported extremely violent shaking • In 55% it was repeated 2 -30 times • In 62% because it stopped the crying • Impact in 24% • Aware of plea bargaining and false confessions BUT for all to be systematic lies defies logic

Story of impact – what forces are required to injure a baby? To kill a baby? Understate the forces “only a jolt / bump” “I sat down hard on the couch with him in my arms” “I put him down on the bed and he bounced a bit” Impact with a soft surface • What might it cause? • What might you find?

Forces - Biomechanical studies • Durhaime – infants cannot be harmed by shaking alone • Primitive models • Injury thresholds for the infant head and brain not been determined (Coats/Margulies) • Infant brains diff from adult – higher water and more vulnerable unmyelinated axons • Bandak • Neck cannot withstand the forces needed to cause AHT • More injuries expected (2004 – neck not routinely Ix)

What about spinal injuries? • Recent studies have suggested that spinal injuries may be more common than previously thought • Injuries may include spinal subdurals, ligamentous injury or spinal fracture • All types of spinal injury more common in AHT than accidental trauma • Tracking, direct impact trauma • We should be routinely imaging the spine Kemp A et al Spinal Injuries in abusive head trauma: patterns and recommendations Pediatr Radiol (2014) 44 (Suppl 4) S 604 -S 612

How can we form an opinion then? • Take each finding separately and exclude alternatives and then put all findings together • Importance of RH’s and spinal findings cannot be over-emphasised eg “SDH could be medical/BESS BUT does not cause RH of this extent and distribution and does not cause spinal haemorrhages/injury” • Controversies bolstered by the legal arena – near complete agreement that shaking a baby is harmful • Diagnosis of exclusion – cannot ever be proven?

TIMING • Imprecise – injury time window • Many factors that can affect the appearance acute SDH (No validated model) • Mixed attenuation do not always suggest repeat episodes • Acute mixed density SDH more common in AHT (hyperacute/acute, blood and CSF – arach tear, acute/chronic haem) • ‘Age different’ lesions strong argument for abuse and may suggest repetitive violence, high risk for further injury

Time of injury How do we tell? Is it • Based on the story provided? • Based on clinical picture (what happens to infant)? • Based on radiological findings? • All of the above? How strong is the evidence base?

Accident versus Abuse The story. . . low height fall, none, minor trauma Factors related to the story (incomplete, changing) Single vs multiple trauma? External vs internal signs of injury? Impact alone (focal) vs complex (widespread) Clinical symptoms – (apnoea, seizures, collapse) Signs (retinal hhges, additional injury, #)

Prevention • Crying as a trigger • Prevention aimed at reducing crying or changing pattern • Changing caregiver response to crying

Summary • AHT poses many challenges to the clinician • Careful consideration of other possible aetiologies is required • High quality evidence exists but the area is also plagued by “non believers!” • Strategies to aid prevention are essential