|Home | About | Journals | Submit | Contact Us | Français|
A torn labial frenum is widely regarded as pathognomonic of abuse.
We systematically reviewed the evidence for this, and to define other intra‐oral injuries found in physical abuse. Nine studies documented abusive torn labial frena in 30 children and 27 were fatally abused: 22 were less than 5 years old. Only a direct blow to the face was substantiated as a mechanism of injury.
Two studies noted accidentally torn labial frena, both from intubation. Abusive intra‐oral injuries were widely distributed to the lips, gums, tongue and palate and included fractures, intrusion and extraction of the dentition, bites and contusions.
Current literature does not support the diagnosis of abuse based on a torn labial frenum in isolation. The intra‐oral hard and soft tissue should be examined in all suspected abuse cases, and a dental opinion sought where abnormalities are found.
Facial and intra‐oral trauma has been described in up to 49% of infants and 38% of toddlers who have been physically abused.1,2 A torn labial frenum (often referred to as frenulum or phrenum) is widely believed by paediatricians to be pathognomonic of abuse,3 and has been described as the most common abusive injury to the mouth.4,5 Several abusive mechanisms have been proposed and include forced feeding,6 gagging, gripping and violent rubbing of, or a direct blow to, the upper lip.7,8 A torn labial frenum, however, is regarded as a trivial intra‐oral injury by dental practitioners, as it is likely to heal spontaneously with minimal complications, and is not reported in large‐scale dental trauma surveys.9
It is estimated that up to 50% of all school‐age children sustain accidental dental injuries,10 challenging clinicians to distinguish between abusive and non‐abusive injuries to the mouth.
We systematically reviewed the literature to establish the probability of a torn labial frenum being caused by physical child abuse and to define what other intra‐oral injuries are found in physical abuse.
We carried out an all‐language literature search of research articles, conference abstracts, websites and references in all articles identified, including review articles and relevant textbooks from 1950 to June 2006. The key words and details of databases searched can be found in intablestables 1 and 22,, respectively. Articles were scanned for duplication and relevance. Authors were contacted where necessary. The resulting studies were reviewed by members of the Welsh Child Protection Systematic Review Group, a panel of 31 people who were paediatricians, child health professionals with child protection expertise, or paediatric or forensic dentists. Each study was subjected to two independent reviews and a third if disputed. Standardised criteria for study definition, data extraction and critical appraisal were used.11 Full critical appraisal and data extraction forms are available at http://www.core‐info.cf.ac.uk.
We included all studies of children aged 0–18 years with intra‐oral injuries due to physical child abuse, and torn labial frena of any aetiology, in live and fatal cases. We defined intra‐oral as the area between the vermilion border of the lips and the hypopharynx.
We excluded review articles, expert opinion or guidelines that did not include primary evidence, studies with mixed adult and child data where the children's data could not be extracted, studies that addressed complications or management of abusive injuries, intra‐oral injuries due to sexual abuse, or thermal injuries (as these will be encompassed in separate reviews) or dental neglect.
Studies were ranked by study design and by the likelihood that abuse had taken place in the “abused” population. We used our own ranking of abuse, as previously described, where a ranking of 1 gave the highest security of diagnosis that abuse had taken place, and a ranking of 5 the least confidence (table 33).12 We included studies with a ranking of 1–4. In the case of non‐abusive injuries, we only included studies where authors had described methods to ensure that abuse had been excluded.
Of 154 studies reviewed, 19 met the inclusion criteria.1,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30 These represented data on 591 children. There were no comparative cross‐sectional or case‐control studies of torn labial frena to enable a probability of abuse to be determined for this injury. The only mechanism described for an abusive torn frenum was a direct blow to the face, recorded in two children. No details of mechanisms were given in other cases. Likewise, there were no comparative studies to compare the characteristics of other abusive and non‐abusive intra‐oral injuries.
Nine studies documented torn labial frena in abused children (table 44):): seven were case series or case studies,14,16,17,18,20,28,29 and one was a case‐control study.23 None of these studies were designed to address torn frena specifically; the case‐control evaluated blunt abdominal trauma in association with cardiopulmonary resuscitation.
These studies represented data on 30 children, of whom 27 (90%) were fatally abused. Twenty two were 5 years of age or younger. The age range given was 0–10 years for five children in one study.23 Four of these studies ranked 1 for abuse, three ranked 3 and two ranked 4.
Cameron was the first to report a torn labial frenum as a consequence of physical abuse in 1966.14 In a retrospective study of 29 fatally abused children between birth and 4.5 years, he noted “the presence in nearly half of the cases of laceration of the mucosa of the inner aspect of the upper lip near the phrenum [sic], sometimes with tearing of the lip from the alveolar margin of the gum was a striking feature of possible significance”.14 Later the same year, Tate29 reported six cases with abusive facial injury, three of whom had a torn frenum. The mechanism was described as a direct blow in two cases: a 2.5‐year‐old child with a torn frenum of the upper and lower lips was shaken and her head struck against a fireplace several times, and a 23‐month‐old child with a torn upper labial frenum was struck about the face. Two of the three children were fatally abused with associated head injuries; multiple bruises and fractures were found in all three.
Of the remaining 11 reported cases, nine were fatally abused with associated head injury16,28,30 (personal correspondence with authors), five had fatal abdominal injuries and one had co‐existent ano‐genital sexual abuse with multiple fractures.18 The remaining two had other intra‐oral injuries, and one later sustained an intracranial injury.17,20
Two separate case studies documented a torn upper labial frenum occurring as a consequence of an intubation, both in fatally injured children.22,23 No further details were offered. When contacted, the authors of the larger series of dental trauma in children21 replied that they did not record a torn frenum as they regarded this as a trivial dental injury.
Fourteen studies documented other types of abusive intra‐oral injuries.1,13,15,17,19,20,21,24,25,26,27,28,29,30 They were all case series or studies, and 11/14 had a higher abuse ranking of 1 or 2. They represented data on 579 children (table 55).). The most commonly recorded abusive injuries to the mouth were lacerations or bruising to the lips. The remaining injuries included mucosal lacerations, dental trauma (including fractures, intrusion and forced extraction), tongue injuries and gingival lesions. No characteristics of these lesions were specific to an abusive aetiology, apart from an adult bite to a child's tongue.19
Becker recorded orofacial trauma in 49% of 260 abused children, 14 (6%) of whom had intra‐oral injuries.1 The largest series, by Naidoo, showed that 59% of 300 physically abused children had facial injuries, 11% of whom had intra‐oral injuries.21 The most common injuries recorded were to the lips (22 children). There were also seven fractured mandibles, and six injuries to oral mucosa, five to teeth, five to gingiva and three to the tongue. Torn frena were not recorded specifically. The children were selected from those with known physical abuse and orofacial injury. No child was examined by a dentist in any of these studies, and the data were collected retrospectively from chart reviews, raising the question of how often the mouth was actually examined.
Injuries to the tongue included an adult bite to an infant's tongue in a 10‐month‐old child, with the arc of the bite pointing towards the lips, confirming it could not have been self inflicted, with multiple fractures, bruises and subdural haemorrhages19 and abrasions and bruising.
In three studies oral bleeding was a presenting symptom.20,26,27 In one case the infant re‐presented five times from the age of 4 months,20 and was found to have a laceration of the uvula with several abrasions and lacerations of the upper gum, hard palate, floor of mouth and lingual frenum; the child developed a retropharyngeal abscess.
Injuries to teeth included displacement, chips, avulsions, intrusion and fractures.1,21,24,25,29,30 One bizarre case series included three siblings who had endured forced dental extraction of permanent incisors as a form of punishment.15
This study confirms that intra‐oral injury occurs in a significant number of children who have been physically abused. Injuries are widely distributed to the lips, gums, tongue and palate and include fractures, intrusion and extraction of the dentition, bruising, lacerations and bites.
There is a paucity of published scientific literature about the torn labial frenum. There are no studies defining the incidence of torn labial frena in abuse and none that compare the injury in abused and non‐abused children. It is impossible, therefore, to ascribe a probability of abuse for a torn labial frenum. Published studies are limited to 30 highly selected cases where a torn labial frenum is described in predominantly pre‐school children who had suffered serious abuse, and where the majority of cases were fatal with extensive associated injuries. The only substantiated cause of an abusive torn labial frenum was a direct blow. There were no recorded cases of forced feeding, twisting or rubbing causing this injury. The literature includes mention of torn labial frena that were not abusive, and therefore a torn frenum in isolation cannot be described as pathognomonic of physical abuse. Clearly the finding of an unexplained torn labial frenum in a young child warrants full investigation, but the paucity of data in the literature and the highly selected nature of cases reported precludes defining an age band where concern would be highest. A diagnosis of physical abuse should not be based on a single injury in isolation, but arrived at in the context of the child's medical, social and developmental history and the explanation offered for the injury.
Some of the largest series in the literature31,32 were not eligible for inclusion as they were only ranked 5 for abuse. They included cases of suspected abuse, with no separation of data on those cases where abuse was actually confirmed.
It is important when assessing a possible torn labial frenum to consider rarer congenital abnormalities of the labial frena such as midline diastema,33,34 hypertrophic frenum in association with hypoplastic left heart syndrome,35 and multiple frena in other congenital heart syndromes such as Ellis‐van Creveld or Pallister‐Hall36,37 as possible explanations of the abnormality.
The most frequently reported abusive injury to the mouth is not a torn labial frenum, as has been suggested,4 but injury to the lips. This assumes that the mouths of all abused children in the various series were fully examined. As these were retrospective studies of case notes,1,21 this is far from clear. None of the described abusive intra‐oral injuries had any diagnostic characteristics, except for a bitten tongue.1,13,15,17,19,20,21,24,25,26,27,28,29,31 It is clear that paediatricians should always examine a child's mouth when assessing a child for suspected abuse. As it is doubtful whether non‐dental specialists would recognise the significance of grey discolouration of the teeth as a micro‐fracture or previous injury,38 or whether they would be able confidently to distinguish this from the characteristic yellow‐brown to grey discoloration found in dentinogenesis imperfecta, it is important to involve dental colleagues. This would be particularly pertinent if co‐existent skeletal fractures were found, as may be the case with combined osteogenesis imperfecta and dentinogenesis imperfecta.39 It is important that paediatricians are aware of the appropriate primary and secondary dentition expected at a given age, in order to question the absence of permanent teeth, as described by Carrotte.15
Many of the intra‐oral injuries described in abused children are likely to be seen by general dental practitioners, yet dentists make very few child protection referrals. This is highlighted by Cotton,40 who noted that of 20000 child abuse investigations only 12 were initiated by dentists. A survey by Becker et al of 1332 dentists in the USA, where there was mandatory reporting with a response rate of 40%, noted that 22 cases of child abuse were seen but only four (18%) were referred to social services,1 despite a legal requirement to report suspected abuse. A similar study by Malecz41 showed that not only did few dentists report abuse cases, but 7% of 155 respondents said that under no circumstances would they report child abuse. Reasons cited included uncertainty about diagnosis and fear of litigation, although practitioners making a referral “in good faith” are protected from litigation in the United States.3 Even 12 years later, a survey of 250 dentists, 157 of whom responded, showed that 50% of dentists had suspected abuse but one third did not refer the case,42 and similar reluctance has recently been documented in the UK where 21% of dentists did not refer cases they suspected of having been abused.43 There have been a number of initiatives in the United States to tackle this issue,44,45 and Welbury et al have developed a computer‐assisted learning programme for general dental practitioners in the United Kingdom,46 and the British Dental Council have recently published guidance.47
There is no legal mandatory reporting of child abuse in the UK, but the British Dental Association has made it clear that dentists do have an ethical responsibility to report child abuse.48 Clearly, those responsible for child protection training in each region must include dental practitioners and hygienists in such programmes, and offer ongoing support.
As in previous reviews,12,49 children with disability were not represented. This is particularly disappointing as disabled children are recorded as being three times more likely to be abused than their able‐bodied counterparts.50
Whilst this review did not deal with dental neglect, a notable number of cases presented co‐existing neglect with resulting dental caries, likely to cause severe pain to the child.25,28 Dental neglect while variably defined,38 should be considered in any child with extensive dental caries or early childhood caries where appropriate dental care has not been sought.51,52 This subject merits a review in its own right, in view of the potential implications in relation to pain, morbidity and faltering growth.53
Future research should be directed at determining the sensitivity and specificity of intra‐oral injuries in abuse by well designed comparative studies. It is clear that we need to define those children who sustain accidental torn frena, in isolation or otherwise, by age, developmental stage and co‐existent injury and cause, in order to aid the distinction from abusive torn frena. Attention should be given specifically to documenting the full extent of intra‐oral injuries in physically abused children and their co‐existent injuries, and this should include disabled children.
The authors wish to thank the following: NSPCC for their financial support of this systematic review; our reviewers: C Adams, M Barber, P Barnes, R Brooks, L Coles, P Davis, R Evans, L Hunter, R Frost, C Graham, M James‐Ellison, B Hunter, R Jenkins, N John, A Kemp, K Kontas, H Lewis, A Maddocks, S Maguire, A Mott, A Naughton, C Norton, M Obaid, M Parry, H Payne, L Price, I Prosser, B Ellaway, J Sibert, E Webb, C Woolley; and Kim Rolfe for technical help with database management and editing of the paper.
Competing interests: None.