Clinical review

When "NAI" means not actually injured

Clifford Mayes, specialist registrar,  Calum Macleod, consultant paediatrician. 

Department of Paediatrics, Antrim Hospital, 45 Bush Road, Antrim BT41 2RL

Correspondence to: Dr Macleod Heather.Houston{at}uh.n-i.nhs.uk

Non-accidental injury is a diagnosis that must be kept in mind by all health professionals who deal with children. The characteristic features of non-accidental injury are well known.¹ We describe a previously unreported case of congenital melanocytic naevi presenting as scalp bruising in an infant with features in the history suggestive of abuse. This case emphasises the need for a cautious, open minded approach to this difficult subject even when the initial history and examination are highly suspicious.

	Case report

Top Case report Discussion References

JD presented to his general practitioner at the age of 3 months with a one week history of bruising to both parietal areas of the scalp. The doctor initially reassured the mother and sent her home. An anonymous telephone call to social services resulted in a second visit that same day with a social worker in attendance, and an urgent referral to a consultant paediatrician was arranged with a suspected diagnosis of non-accidental injury. The mother was insistent that she had not harmed the child in any way and was clearly distraught that such an allegation had been made. In hospital the mother stated that the bruises had been caused by the infant rolling his head from side to side in an unpadded car seat. She could not, however, describe a specific event and explained her delay in seeking medical advice by saying that she had asked her own mother for advice and had been reassured by her. JD was the product of a normal pregnancy and delivery and had an otherwise unremarkable medical history. He was the only child of an unmarried mother who cohabited with the child's father.

On examination he was well cared for. His growth parameters were all just above 50th centiles. Three flat, non-tender, brown lesions resembling bruises were readily apparent on both parietal areas of the scalp: on the right a single, well circumscribed lesion measuring 5.0×4.5 cm; on the left two discrete, linear lesions measuring 4.5×1.0 cm and 3.5×0.75 cm respectively. It was noted that the hair overlying the lesions was darker than elsewhere. The remainder of the examination was normal. In particular there was no bruising elsewhere, the frenulum was intact, and fundi were normal. Interaction between mother and child seemed appropriate.

Admission to hospital was arranged in view of the physical findings of bruising to the head of a premobile infant, the inadequacy of the explanation (inspection of the car seat showed that it could not possibly be the mechanism for his injuries), the delay in seeking medical advice, and the concerns of social services, which were otherwise considering an emergency protection order. The mother understood the need for admission and gave her verbal consent, although she later stated that she felt she had no choice but to comply.

Inpatient investigation revealed a normal cranial ultrasound scan, skeletal survey, full blood picture, and coagulation screen. Serial examination of the lesions showed no change over a period of four days in hospital, and an alternative diagnosis was therefore considered. The opinion of a consultant dermatologist was sought, and this confirmed the diagnosis of congenital melanocytic naevi. A biopsy of the lesions was not considered necessary as the clinical appearance by this stage was clear and subjecting the infant and his parents to further investigation was considered inappropriate. JD was discharged to the care of his parents with plans for review in a dermatology clinic.

	Discussion

Top Case report Discussion References

We describe a previously unreported example of a benign congenital skin lesion presenting as physical abuse. A Medline search revealed no prior reports of this in the literature, and our case adds to those cutaneous lesions that have previously been confused with non-accidental injury.^2-4

Cutaneous lesions are the most obvious and common manifestation of child abuse, with bruises being the commonest such lesion.⁵ Furthermore, the presence of bruising on the head and neck of an infant is strongly associated with a non-accidental cause.⁶ In this case there had also been a delay in seeking medical advice as well as a vague and inconsistent explanation for the apparent injury, features held to be characteristic of non-accidental injury. All these factors required a full assessment and investigation, which happily led to the correct diagnosis being made and the infant returning to the family home.

In retrospect we can see that the explanation for the lesions, which actually made the professionals involved more suspicious, was the result of a concerned mother trying to explain what she saw rather than hide what she had caused. A question that arises is why the lesions should have become apparent only at 3 months of age. The likely explanations are that the natural course of congenital melanocytic naevi is to darken with age and that the hair of the newborn infants tends to thin over a period of months, thus revealing the lesions.

The social consequences of this admission should not be underestimated. It would seem likely that it was a neighbour who telephoned social services. During the hospital stay, the mother talked of whispering neighbours and rumours of JD having been admitted to intensive care. Given that the family lives in a small rural community such comments were clearly a source of considerable additional distress. It took four days for the doctors and social workers involved to be convinced that the infant had not been physically abused, but it may take considerably longer for some members of the local community to be similarly convinced.

Awareness of non-accidental injury among healthcare professionals and the public has markedly increased since the first description over 30 years ago.⁷ This case illustrates the difficulties inherent in making the diagnosis. The consequences of failing to diagnose child abuse can be catastrophic, but it is also essential that we are able to recognise benign conditions that may masquerade as abuse if we are to avoid unnecessary distress. Even in retrospect, it is difficult to formulate a different management plan for this infant.⁸ This case clearly illustrates that apparently classic features do not necessarily add up to a diagnosis. In a case that initially seems quite clear it is essential that an open mind and an empathetic approach be maintained throughout.

	Acknowledgments

We acknowledge the assistance of Dr H Jenkinson in the management of this patient.

	Footnotes

Classic      features of non-accidental injury are not always diagnostic

	References

Top Case report Discussion References

1. Meadow R. ABC of child abuse. London: BMJ Publishing , 1983.
2. Alder R, Kane-Nussen B. Erythema multiforme: confusion with child battering syndrome. Pediatrics 1983; 72: 718-720[Abstract/Free Full Text].
3. Brown J, Melinkovich P. Schonlein-Henoch purpura misdiagnosed as suspected child abuse: a case report and literature review. JAMA 1986; 256: 617-618[Medline].
4. Gellis S, Feingold M. Cao gio: pseudobattering in Vietnamese children. Am J Dis Child 1979; 130: 856-858.
5. Ellerstein NS. The cutaneous manifestations of child abuse and neglect. Am J Dis Child 1979; 133: 906-909[Abstract].
6. Hobbs CJ, Hanks HG, Wynne JM. Child abuse and neglect, a clinician's handbook. Edinburgh: Churchill Livingstone , 1993.
7. Kempe CH, Silverman FN, Steele BF, Droegmueller W, Silver HK. The battered child syndrome. JAMA 1962; 181: 17-24.
8. American Academy of Pediatrics, Committee on Hospital Care, Committee on Child Abuse and Neglect. Medical necessity for the hospitalisation of the abused and neglected child. Pediatrics 1998; 101: 715-716[Abstract/Free Full Text].

(Accepted 23 October 1998)