Rapid Responses to:

RESEARCH: Alison M Kemp, Frank Dunstan, Sara Harrison, Susan Morris, Mala Mann, Kim Rolfe, Shalini Datta, D Phillip Thomas, Jonathan R Sibert, and Sabine Maguire Patterns of skeletal fractures in child abuse: systematic review BMJ 2008; 337: a1518 [Abstract] [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Mandatory Osteogenesis Imperfecta Screening in Suspected Non-accidental Injury?

Chika Uzoigwe, F Shabani   (12 October 2008)

"Unexplained" fractures

James R Le Fanu, Rosemary Neary, Denise Bartlett   (18 October 2008)

unexplained fractures

Alison M Kemp, Frank Dunstan, Sara Harrison, Susan Morris, Mala Mann, Shalini Datta, Phillip Thomas, Jo Sibert, Sabine Maguire   (18 November 2008)

Mandatory Osteogenesis Imperfecta Screening in Suspected Non-accidental Injury? 12 October 2008
Chika Uzoigwe, Orthopaedics Sheffield, F Shabani Send response to journal: Re: Mandatory Osteogenesis Imperfecta Screening in Suspected Non-accidental Injury?	It is important not to overlook osteogenesis imperfecta (OI) in children presenting with recurrent or unusual fractures. Osteogenesis imperfecta represents a spectrum of disorders; some of which result in a slight but significance increase in fracture risk1. Milder forms can present later as children become more active or become more frequently involved in “horse-play” with siblings or peers in infant school. There may be no overt family history of the diseases2. In such cases recurrent fractures or healed old fractures may be the presenting feature. Clinicians must be alert to the clinical features of OI including discolouration of the sclera, short stature, skeletal deformity and dentogenesis imperfecta. Affected children may also appear unremarkable. Biochemical diagnosis is achieved by analysis of procollagen derived from skin fribroblasts1. Marlowe and co-workers tested for OI in a cohort of children where non- accidental injury (NAI) was suspected. They identified OI in 11 of the 138 cases where there was sufficient clinical information to make an evaluation. In a further 11 OI could not be excluded3. False accusation of child abuse can destroy lives and families. However, it is important to mention that, statistically, NAI is more common than OI2. In all cases of children presenting with atypical fractures, non- accidental injury must be considered. However the possibility of OI should also be entertained. There are a number of cases where OI has been mistaken for NAI, which have led to criminal prosecution and children being separated from their families3. The consequences of the diagnosis of NAI are so serious that there is an argument for collagen testing, in cases where the basis of suspicion is skeletal injury. 1. Rauch F, Glorieux FH. Osteogenesis imperfecta. Lancet. 2004; 24; 363:1377-1385 2. Marlowe, M G Pepin, P H Byers Testing for osteogenesis imperfecta in cases of suspected non-accidental injury. J Med Genet. 2002; 39:382-6. 3. Paterson CR, McAllion SJ. Classical osteogenesis imperfecta and allegations of non-accidental injury. Clin Orthop Relat Res. 2006; 452:260 -264. Competing interests: None declared
"Unexplained" fractures 18 October 2008
James R Le Fanu, General Practitioner Mawbey Brough health Centre, 39 Wilcox Close, London SW8 2UD, Rosemary Neary, Denise Bartlett Send response to journal: Re: "Unexplained" fractures	15 October 2008 The Editor British Medical Journal Dear Editor Both the systematic review by Kemp and colleagues of skeletal fractures in children and the accompanying Editorial (1,2) gloss over the substantial medical uncertainties in this area that result in parents being wrongfully accused of injuring their children. It might seem self evident that deliberately inflicted injury must be the probable cause of an ‘unexplained’ fracture - where ‘unexplained’ in this context means there is no plausible history of accidental trauma and those conditions that predispose to bone fragility, such as osteogenesis imperfecta, have been excluded. The uncertainty arises within that unexplained category where there is a marked discrepancy between the clinical presentation of the child and purported abusive nature of the fracture: where he appears well cared for by responsible parents and with no circumstantial evidence in the form of bruising or soft tissue swelling that would be expected were he the victim of physical assault. This discrepancy is particularly apparent in our experience as a parent support group in the more than one hundred cases with which we have been involved over the past five years where a skeletal survey following an apparently spontaneous fracture reveals further multiple, symmetrical metaphyseal fractures especially around the knees and ankles. The prevailing view amongst paediatric radiologists is that these fractures are characteristic of abusive injury induced either by violent shaking or ‘direct wrenching or twisting of the limbs’. (3) It must, however, be staggeringly unlikely on clinical and commonsensical grounds to suppose that apparently responsible parents should deliberately shake or twist the limbs of their babies so as to cause the same pattern of fractures – but in such a way as to leave no external stigma of physical assault. Rather, it is reasonable to presume the probable cause of this pattern of fractures, that significantly only occurs in the first six months of life, is a developmental bone disorder of as yet unknown (or overlooked) cause. (4) Regrettably, paediatricians in this situation tend to defer to the historically legitimated opinion of paediatric radiologists that these fractures are indicative of abuse – with predictably devastating consequences for parents and their families. Yours sincerely Rosemary Neary, Denise Bartlett, The Eaton Foundation, Wisbech, Cambs PE14 8TP James Le Fanu MRCP, Mawbey Brough Health Centre, 39 Wilcox Close, London SW8 2UD References 1. Kemp A M, Dunstan F, Harrison S et al; Welsh Child Protection Systematic Review Group. Is this fracture due to abuse? A systematic review of the patterns of skeletal fractures in child abuse. BMJ 2008; 337; pp859-862 2. Naomi F Sugar; Diagnosing Child Abuse. BMJ 2008; 337; pp826-827 3. Kleinman P, Marcs S C, Blackbourne B. The Metaphyseal Lesion in Abused Infants: A radiologic-histopathologic study. Am J Radiol 1986; 146; 895-905 4. Marvin Miller; The Lesson of Temporary Brittle Bone Disease: all bones are not created equal. Bone 2003; 433; pp466-474 Competing interests: RN, DB, JLF, are involved in the work of the Eaton Foundation for Parents Accused of NAI
unexplained fractures 18 November 2008
Alison M Kemp, Reader in Child Health Cardiff University, Frank Dunstan, Sara Harrison, Susan Morris, Mala Mann, Shalini Datta, Phillip Thomas, Jo Sibert, Sabine Maguire Send response to journal: Re: unexplained fractures	In response to the letter from Dr Le Fanu et al BMJ 2008;337. We would have hoped that our lengthy description and discussion of the quality of literature in this field would have made clear the benefits and deficiencies of current scientific publications 1. The systematic review addressed the question “What features differentiate abusive fractures from those due to other causes” and concludes that • a proportion of children who have been physically abused do sustain skeletal fractures • abusive fractures are most prevalent in children under eighteen months of age • abusive fractures may be multiple and can be widely distributed throughout the skeletal system • there is no specific fracture that on its own is diagnostic of abuse • there are deficiencies in the scientific literature when comparing certain fracture types. This is particularly notable with regard to comparative studies of metaphyseal fractures “When are parents wrongfully accused of injuring their children” is a separate question and requires a separate systematic review of the published literature with different search criteria. In the course of this review, we identified cross sectional studies that identified differential causes of fractures that varied dependant upon the age group of children and fracture type being studied. Broadly the causes of fractures in the infant and toddler age group included major trauma, falls, conditions that predispose to bone fragility (osteopenia of prematurity, bone dysplasia, osteogenesis imperfecta and metabolic conditions) as well as physical child abuse. We would therefore re-iterate our recommendations that clinicians should “consider child abuse as part of a differential diagnosis” in this age group. A standard diagnostic approach must consider all possible causes of unexplained fractures and incorporate appropriate history, examination, biochemical and radiological investigation and interpretation 2. We were disappointed that despite a thorough search of over fifty years of international literature we could find no published cross sectional studies or comparative studies of metaphyseal fractures in children and would agree that this is a necessary piece of future research. As stated in our review, a handful of studies show that metaphyseal fractures are more common though not exclusively seen in child abuse. There are undoubtedly case reports and case series that show that metaphyseal fractures have been associated with severe cases of child abuse 3-7 and case reports of metaphyseal fractures in correctional talipes procedures 8 and in birth related injuries 9. Le Fanu et al assert that abusive fractures would be accompanied by “bruising or soft tissue swelling that would be expected were he or she the victim of physical assault” and express doubt that apparently responsible parents should deliberately shake or twist the limbs of their babies so as to cause the same pattern of fractures—but in such a way as to leave no external stigma of physical assault” . It is clearly established, and clinically entirely plausible that fractures are not always accompanied by bruises, Merten et al have shown that 80% of rib fractures were clinically unsuspected at diagnosis 10. In addition, abused children may suffer multiple fractures of different ages; clearly a fracture which is some 6-8 weeks old is even less likely to be accompanied by bruising or soft tissue swelling. The nature of certain fractures (rib fractures, metaphyseal fractures) makes external bruising on the skin even more unlikely. In the case of abusive rib fractures, the point of fracture is described as being on the inner surface of the rib, and could not therefore be expected to show external bruising to the overlying skin. In relation to metaphyseal fractures, Kleinman’s studies showed that acute fractures were difficult to visualise on skeletal survey, at post mortem intramedullary haemorrhage and evidence of subpeniosteal bleeding or visible hemorrhage were seldom found on histological examination 5. This again confirms that external bruising and soft tissue swelling are unlikely. The absence of bruising in fractures is well recognised in Accident and Emergency practice and many subtle fractures in children present without soft tissue swelling. In relation to hypothesised conditions where “the “probable” cause is an unknown, undefined condition which cannot be tested for, and has no clinical studies to characterise it", it is impossible to ascertain what fractures may or may not be associated with such an ill defined entity. We are disappointed that Dr Le Fanu does not recognise that our systematic review can only critically appraise the world literature relating to primary studies of fractures with known causes, and our methodology is explicit in determining which fractures are included in our meta-analysis dependent on confirmed aetiology. Clearly there is scope for future research to define further the specific probabilities of certain fractures, a point that we made at the end of our review. We look forward to further scientific contributions, by high quality primary studies, to add to our understanding of the nature and characteristics of fractures that children sustain by inflicted and non-inflicted means. 1. Kemp A M, Dunstan F, Harrison S, Morris S, Mann M, Rolfe K, et al. Is this fracture due to abuse? A systematic review of the patterns of skeletal fractures in child abuse. BMJ 2008;337 2. Bishop N, Sprigg A, Dalton A. Unexplained fractures in infancy: looking for fragile bones.b Archives of Disease in Childhood 2007;92:251-256; 3. Kleinman, P. K., & Marks, S. C., Jr. (1996a). A regional approach to classic metaphyseal lesions in abused infants: the distal tibia. American Journal of Roentgenology, 166(5), 1207-1212 4.Kleinman, P. K., & Marks, S. C., Jr. (1996b). A regional approach to the classic metaphyseal lesion in abused infants: the proximal humerus. American Journal of Roentgenology, 167(6), 1399-1403 5. Kleinman, P. K., & Marks, S. C., Jr. (1996c). A regional approach to the classic metaphyseal lesion in abused infants: the proximal tibia. American Journal of Roentgenology, 166(2), 421-426 6. Kleinman, P. K., & Marks, S. C., Jr. (1998). A regional approach to the classic metaphyseal lesion in abused infants: the distal femur. American Journal of Roentgenology, 170(1), 43-47 7. Kleinman, P. K., Marks, S. C., Jr., Richmond, J. M., & Blackbourne, B. D. (1995). Inflicted skeletal injury: A postmortem radiologic- histopathologic study in 31 infants. American Journal of Roentgenology, 165(3), 647-650 8. Grayev, A. M., Boal, D. K. B., Wallach, D. M., & Segal, L. S. (2001). Metaphyseal fractures mimicking abuse during treatment for clubfoot. Pediatric Radiology, 31(8), 559-563 9. Altman, D. H., & Smith, R. L. (1960). Unrecognised trauma in infants and children. Journal of Bone and Joint Surgery, 42(3), 407-413 10 Merten DF, Radlowski MA, Leonidas JC. The abused child: A radiological reappraisal. Radiology. 1983 Feb;146(2):377-81. Competing interests: Authors of original article