A study of the validity of a simulation of third molar eruption based on Markov modelling

PURPOSE OF INVESTIGATION

The study was undertaken to examine the extent to which a model based on a Markov modelling process would simulate the eruption of lower third molars.

BASIC PROCEDURE

A Markov process model was developed to model eruption of lower third molars using a 1-year time increment based on simulated eruption data relating to 100 lower third molars. This model was tested using a Monte Carlo simulation and compared with eruption patterns of 973 actual lower third molars. Statistical analysis of the differences between the simulated and actual groups was undertaken.

MAIN FINDINGS

There were no differences between the simulated and actual data other than for the subgroup aged 30-34 where the simulation overestimated the probability of parteruption and underestimated the probability of uneruption and full eruption.

PRINCIPAL CONCLUSIONS

This methodology produced an accurate model that may be of use in service planning.

Intra-observer reliability regarding removal of asymptomatic third molars

OBJECTIVE

To investigate reliability of practitioners' removal decisions and judgements of risk of pathology associated with asymptomatic third molars.

SUBJECTS

10 oral surgeons and 18 family dentists from South Wales with experience ranging from 5 to 28 years.

METHOD

Participants were presented with periapical radiographs of 36 asymptomatic, mandibular third molars and were informed of the age and sex of the patients and the degree of eruption of the third molars. Participants were asked to assess, using visual analogue scales, the likelihood of future pathology if the third molars were left in situ and to indicate if they should be removed or not. To assess intra-observer reliability, the 36 cases were duplicated and presented to the participants on a different occasion, a month later. The same questions were asked as on the first occasion.

RESULTS

Significant correlations (Pearsons correlation coefficients) were found between initial and repeat assessments of all measures but there was little agreement about the need for removal (Kappa values: 0.54 for oral surgeons and 0.41 for the family dentists). For every item studied, changes in position on the visual analogue scale of two-thirds or more of the total length occurred from the first to the second assessment.

CONCLUSION

Treatment decisions about whether or not to remove asymptomatic third molars were not made on a rational basis. Since similar conclusions were recorded in a previous Swedish study, it is inferred that until further high quality evidence of disease prediction is published, decisions to remove third molars prophylactically cannot be made reliably.

Neural networks: a new technique for development of decision support systems in dentistry

OBJECTIVES

To outline the key concepts of neural network based systems and to evaluate the potential applications of such systems in dentistry.

DATA SOURCES

Published work on neural networks.

CONCLUSIONS

Neural networks may initially seem complex and computer intensive, but actually integrate well with a clinical environment. Neural network expert systems may be trained with only clinical data and as such can be used where 'rule based' decision making is not possible. This is the case in many clinical situations. Neural networks may therefore become important decision making tools within dentistry and have applications both in improving clinical care and in maximizing the cost benefit of care.

The BAOMS United Kingdom survey of facial injuries part 1: aetiology and the association with alcohol consumption. British Association of Oral and Maxillofacial Surgeons

OBJECTIVE

To determine the age and sex distribution, timing, causes, geographical location, and nature of facial injuries in the UK and to determine the association of these factors with alcohol consumption by the patient or any other involved person.

DESIGN

A 12-section proforma was completed on all patients with facial injuries covering their age and sex, time and day of injury and presentation, the cause and type of injury and where it occurred, the treatment the patient received, any other injuries, and alcohol consumption by the patient and any other involved person. The total attendances for the study week and the catchment population for each A&E department were recorded.

SETTING

163 of the A&E departments in the UK served by 137 of the UK's oral and maxillofacial departments.

SUBJECTS

All patients who presented with facial injuries to these 163 A&E departments in England, Scotland, Wales and Northern Ireland over the study week from 09.00 hours on Friday 12 September 1997 to 08.59 hours on Friday 19th September 1997.

RESULTS

6114 patients with facial injuries presented over the week, out of a total of 152,692 A&E attendances. The male:female ratio was 68:32. This rose to 79:21 in assault cases. The mean age of all patients was 25.3 years, of males 23.2 years, and of females 29.8 years. Forty per cent of the facial injuries were caused by falls. A large proportion of these happened to the under-5 age group in the home. Eleven per cent of all falls were associated with alcohol consumption. Twenty-four per cent of the facial injuries were caused by assault. The commonest sites for assault were the street followed by public drinking establishments. More women than men were assaulted at home. Fifty-five per cent of assaults were related to alcohol consumption. Eight per cent of assaults were with bottles or glasses. Five per cent of the facial injuries occurred in road traffic accidents (RTAs). Fifteen per cent of RTA victims had consumed alcohol. The 15-25 age group suffered the greatest number of facial injuries caused by assault and RTAs and had the highest number associated with alcohol consumption. At least 22% of all the facial injuries in all age groups were related to alcohol consumption within 4 hours of the injury. In the over 15 age groups, alcohol consumption was associated with 90% of facial injuries occurring in bars, 45% on the street, and 25% in the home. Assault, RTA and alcohol consumption conveyed an increased risk of serious facial injury.

CONCLUSIONS

Campaigns should be instituted to educate young people about the link between excessive alcohol consumption, assault, road traffic accidents and serious facial injury.

Vulnerability, victims and violence

The burden on accident and emergency (A&E) departments of dealing with the aftermath of violence has increased substantially in the last 10 years. Both the underlying causes and the effects on the injured are multifaceted. It is important that clinicians who treat the injured know which interventions, like early family support and preschool education, are effective in preventing violence. It is also important to target those who will benefit from interventions to prevent, for example, serious psychological sequelae or further alcohol related harm. As with child protection, the organisation of services for adults needs to be built on teamwork with other agencies, particularly so that the injured are assessed for risk of future harm and are given opportunities to report offences to the police if they wish. Up to now, the contribution of A&E doctors to dealing with violence has been largely reactive. This policy paper argues that since many of the underlying causes and circumstances of violence can be modified, a more proactive, interagency approach would be effective in the fight against violence.

An investigation of preferential fibroblast wound repopulation using a novel in vitro wound model

To overcome the difficulties of studying wounding and wound repopulation in monolayer systems, a 3-dimensional model of wound repopulation has been developed which allows the in vitro investigation of fibroblast migration in response to experimental wounding. This model was utilized to determine whether fibroblasts derived from sites which demonstrate preferential healing (child and oral mucosal fibroblasts) possessed an increased ability to repopulate experimental wounds when compared to adult dermal fibroblasts. Fibroblasts were established from specimens derived from healthy donors undergoing minor elective surgery. Standard wounds were created in fibroblast populated collagen lattices (FPCLs) which were then overlaid upon an extracellular wound matrix. Fibroblast repopulation of the wounds was studied over 12 days using light- and scanning electron microscopy and quantified using computerized image analysis. Wound repopulation by fibroblasts derived from child donors (n = 3) was significantly (P < 0.001) more rapid than their adult tissue-matched counterparts (n = 3). Wound repopulation by oral mucosal fibroblasts (n = 3) was significantly greater than that exhibited by age-matched dermal fibroblasts (n = 3; P < 0.05). These differences were not reflected in differences in DNA synthesis (P > 0.5) or cell number (P > 0.5) within similar attached FPCL systems. These findings further support the concept of a gradual transition from the fetal to adult phenotype in wound healing. The potential applications of the model are discussed.

Psychological sequelae of facial trauma

BACKGROUND

Facial trauma is common in accidents and assaults and can be accompanied by distressing psychological sequelae.

METHODS

Retrospective analysis of case notes followed by a prospective assessment of consecutive facial trauma victims was performed to determine the prevalence and detection rate of psychological sequelae.

RESULTS

Only 8 of 47 case notes (17%) contained any reference to mental state. Forty-three patients completed initial assessment and 7-week questionnaires. Twelve (27%) were suffering from posttraumatic stress disorder at 7 weeks. Factors significantly associated with poorer outcome were higher initial Hospital Anxiety and Depression Scale and Impact of Event Scale scores, assault injury fractures, and prediction of psychological sequelae by junior oral surgeons.

CONCLUSION

More than one-quarter of victims of facial trauma in this study developed posttraumatic stress disorder, documentation of which was poor. The study suggests that psychological sequelae can be predicted using simple measures such as a basic initial assessment by nonmedically qualified personnel.

Agreement between practitioners concerning removal of asymptomatic third molars

OBJECTIVE

To investigate and compare agreement within two groups of dental practitioners, family dentists and oral surgeons, in their decisions regarding removal of asymptomatic mandibular third molars.

SUBJECTS

Ten oral surgeons and 18 family dentists from South Wales with experience ranging from 5 to 28 years.

METHODOLOGY

Participants were presented with periapical radiographs of 36 asymptomatic, mandibular third molars and were informed of the age and gender of the patients and the degree of eruption of the third molars. Participants were asked to indicate whether they thought that the third molar should be removed or not. The degree of agreement between participants was measured by kappa indices for multiple raters.

RESULTS

The kappa indices were 0.14 for the oral surgeons and 0.09 for the family dentists, indicating poor agreement beyond chance. Although in most cases the participants decided not to remove the third molar, they did so inconsistently, that is, they did not make this decision on the same cases. There were also differences in the inclination of the participants to suggest removal of the 36 third molars.

CONCLUSION

Poor inter-observer agreement suggested that treatment decisions regarding asymptomatic third molars are based more on subjective beliefs and habitual practices than on rational decision making.

Assessment of repair of facial lacerations

A survey of senior nurses in 100 randomly selected large (<30000 new patients/year) UK Accident & Emergency Departments (AED) was undertaken to investigate which grade of staff/specialty sutured facial lacerations and to obtain an inter-specialty perspective of standards of care. Of the 80 hospitals whose AED senior nurse responded, half had resident oral and maxillofacial surgeons but only 20% had resident plastic surgeons. Lacerations were mainly sutured by AED doctors (55% of hospitals), oral and maxillofacial surgeons (21%), plastic surgeons (12%) and AED nurses (11%). Overall the highest standard of care was considered to be provided by oral and maxillofacial and plastic surgery registrars and the lowest by nurses and consultants. Overall, oral and maxillofacial surgery services were ranked first, plastic surgery second and A&E third. Monofilament polypamide was the most favoured suture material, but polybutester and polypropylene were also widely used.

Facial protection conferred by cycle safety helmets: use of digitized image processing to develop a new nondestructive test

Cycle safety helmets are designed to prevent head injury. Although most commercially available helmets conform to one of several national and international standards, individual designs differ widely, particularly in relation to face coverage. A method was developed to assess the potential for the differing designs to protect the face from injury. A nonimpact test was assessed, using digitized image-processing software (Digithurst Ltd.) to measure the shadow cast by a helmet rim under a collimated plane light source onto the face of a mannequin headform. Twelve helmet designs available internationally were tested and ranked with respect to the direct protection conferred (area of the face directly covered by the helmet) and indirect protection (area of the face shaded). The three highest-ranking helmets for direct protection (Rosebank Stackhat, Asphalt Warrior, and Lazer Voyager) also ranked the highest for indirect protection. These helmets were more inferiorly extended and were of a more bulky construction. It was concluded that the dimensions of cycle helmets in relation to face coverage are crucial in influencing the extent to which facial protection is conferred. International test standards need urgent revision to ensure that face coverage is optimized. Lower-face protection could be achieved through incorporation of a lower-face bar to cycle helmets.

The appropriateness of referral of medically compromised dental patients to hospital

Hospital departments of oral and maxillofacial surgery make a substantial contribution to both managing and treating medically-compromised dental patients. Contracting arrangements should take account of this. Demographic data suggest that the treatment of medically-compromised elderly dentate patients will become increasingly important in the General Dental Service (GDS). To determine the medical conditions and treatment requirements prompting referral of these patients to hospital, a prospective study was undertaken of 75 consecutive adults referred for hospital treatment specifically because of a medical condition which prevented delivery of routine dental care in the GDS. Patients (mean age: 56 years) were referred mainly from general medical (33%) and dental (62%) practitioners. Cardiovascular disease was the most frequently cited medical condition requiring referral (43%; n = 32 cases). Forty-eight patients (64%) were symptomatic on presentation and on average had attended on 2.3 occasions before definitive treatment was instituted. Fifty-two patients (70%) had no special treatment requirements other than those available in the GDS, 11 patients (15%) simply required antibiotic prophylaxis and 81% were treated by undergraduates or junior staff. These data suggest that many patients referred for dental hospital treatment because of underlying medical condition are not in fact medically-compromised and may be treated in the primary care setting.

Risks associated with occupational glass injury in bar staff with special consideration of hepatitis B infection

Since bar workers often sustain cuts from unwashed bar glasses, the aims of this study were to investigate risk of injury and to examine the sero-prevalence of markers for hepatitis B amongst bar staff. Ninety-one bar staff recruited by newspaper advertisement were asked about injury experience and life-style risks associated with transmission of hepatitis B and were tested for hepatitis B surface antigen (HBsAg) and core antibody (anti-HBc). Seventy-four per cent reported lacerations from broken glassware at work: 18% had sustained such injuries in more than 10 incidents. Fifty-five per cent of respondents reported occupational skin contact with body fluids. Anti-HBc prevalence for the study group was 1.1%, suggesting that bar staff were not at increased risk from hepatitis B infection. Although 30% wore gloves for high-risk tasks, there was no evidence that glove wearing prevented glass lacerations. This level of injury experience and exposure to body fluids is unacceptable and represents a potential risk of cross-infection. Hepatitis B immunization should be considered in this group. Urgent action, including the replacement, wherever possible, of annealed with tempered bar-glassware, is necessary to protect bar workers from glass injury.

Comparisons of the abilities of a neural network and three consultant oral surgeons to make decisions about third molar removal

AIM

To compare the performance of a computer based decision support system (a neural network) and consultant oral and maxillofacial surgeons in making decisions about the need to remove lower third molars.

DESIGN AND SETTING

Receiver operating characteristic (ROC) analysis at a hospital department of oral and maxillofacial surgery.

SUBJECTS AND METHODS

Three consultant oral and maxillofacial surgeons indicated on a six-point rating scale how certain they were that each of 50 documented lower third molars required removal. Similar data were obtained from the neural network following appropriate coding of the clinical information. These data were compared with gold standard treatment decisions for each tooth based on National Institutes of Health Concensus criteria using ROC analysis.

MAIN OUTCOME MEASURES

The area beneath each operator ROC curve (varying between zero and one with greater areas indicating better performance).

RESULTS

The network performed as well a two consultants (P = 0.12/0.18, NS) and significantly better than the third (z = 526, P < 0.01).

CONCLUSIONS

This work suggests that this computer based neural network could play a useful role in supporting dental practitioners making third molar referral decisions.

Trends in the referral and treatment of new patients at a free emergency dental clinic since 1989

AIM

To investigate the changes in the pattern of new patient referral and treatment at a dental hospital emergency clinic since the introduction of the new dental contract in 1989.

DESIGN

A prospective survey and review of clinical records.

SETTING

The examination and emergency clinic at Cardiff Dental Hospital.

SUBJECTS

500 consecutive new patients attending the clinic in May/June in 1989, 1993, 1994 and 1995.

MAIN OUTCOME MEASURES

Numbers of referrals; source of referral; geographical distribution of patients; registration with general dental practitioners; type of dental disease; treatment received.

RESULTS

New patient attendances increased by 56% since 1989. Referrals from family doctors and dentists decreased while self-referrals increased. Self-referral by patients not registered with family dentists increased by 56%. Extraction of teeth was the commonest form of treatment in all years and increased by 72%.

CONCLUSIONS

Numbers of patients attending for dental treatment at this setting increased substantially after 1989. Increasing numbers of these patients were inappropriately self-referred for treatment. Patients increasingly selected extraction of teeth as the primary treatment rather than restorative procedures. These changes appeared to be unrelated both to the severity of dental disease at presentation and to socio-economic status.

Multiattribute utility assessment of outcomes of treatment for head and neck cancer

Good clinical practice is dependent on continuous audit. Most audits of head and neck cancer treatment planning have been subjective, with only 5-year survival rates being considered objectively. Improvements in clinical care require not only measurable goals that relate to patients' perspectives, but also a means of assessing to what extent those goals have been met. In this context, 5-year survival rates are too crude to be useful, although they remain important for other reasons. Because a simple clinical objective measure of outcome applicable to head and neck cancer is not available, multiattribute assessment techniques were used to develop a clinically based scale for outcomes following treatment for head and neck cancer, with domains centred on social function, pain, physical appearance, eating and speech problems, nausea, donor site problems and shoulder function. Domains were weighted relative to each other; pain (mean weight 85) and social function (89) were considered most important followed by physical appearance (76), eating (76) and speech problems (74) A series of graded statements was constructed within each domain and scaled relative to each other. These components were also combined into an overall scale that will enable objective outcome assessment in this important area of medical care.

Lower third molar treatment planning ability of general dental practitioners and oral maxillofacial surgeons using receiver operating characteristics methodology

A comparison of lower third molar treatment planning ability of six oral and maxillofacial surgeons (three consultants, two senior registrars and one senior house officer) and 27 general dental practitioners was carried out using Receiver Operating Characteristics (ROC) methodology. Each clinician was presented with the same series of 50 case histories and asked to indicate, using a rating scale, how certain they were that each lower third molar required removal. These data, together with NIH gold standard treatment decisions for each third molar, were then used to construct ROC curves for each clinician and, using combined data, to produce ROC curves for the groups of hospital specialists and general dental practitioners. These curves were then compared statistically. There were significant differences between the best and worst practitioners, both in the hospital service and within general dental practice. When combined data were examined, however, there was no statistical difference in the treatment planning ability of hospital specialists and general dental practitioners. These findings indicate that lower third molar treatment planning ability is more dependent upon individual factors than specialist training. Further, the data presented here suggest that general dental practitioners are as capable of treatment planning for lower third molars as the specialists to whom they refer.

Antibiotic prescription for acute dental conditions in the primary care setting

To determine the pattern of antibiotic prescribing (source and type) for acute dental conditions in the emergency setting, a prospective cross-sectional study was undertaken of 500 consecutive new patients attending the examination and emergency clinic of the Cardiff Dental Hospital in May 1994. The source and type of antibiotics prescribed, the nature of complaint and type of antibiotic prescribed in the primary care and dental hospital setting were recorded. Antibiotic prescription was a feature of the treatment of 30% (149/500) of patients. The 60 patients who had received emergency treatment prior to attending the hospital emergency clinic had been prescribed a total of 14 different types of antibiotics; 41% of these patients had sought treatment from general medical practitioners. Antibiotics were frequently prescribed without generally accepted criteria and there was wide variation in prescribing. Overall, only seven different types of antibiotics were prescribed in the dental hospital setting. The use of second generation antibiotics was more common in general practice than in the dental hospital. The results suggest that many patients with dental pain may seek treatment from medical practitioners, prior to, or in place of, definitive dental treatment. In summary, rationalisation of antibiotic prescription and the provision of emergency dental treatment is needed.

Prevalence of third molars in dental practice attenders aged over 35 years

There is a lack of information in the United Kingdom on the prevalence of third molars in older patients. The aim of this study was therefore to define the pattern of lower third molar retention in UK dental practice attenders aged 35 years and over. A random sample of 599 eligible patients from a rural dental practice were included in the study. Information was obtained from clinical notes and panoramic radiographs. Data collected included age, gender, presence or absence of lower third molars, number of teeth present in the lower arch and eruption status. Two hundred and sixty-four (44.1 per cent) had at least one lower third molar present (mean age = 57.1 years) while 335 (55.9 per cent) had no lower third molars (mean age = 50.2 years). The data suggest that a greater proportion of men than women retain at least one lower third molar although this finding was not statistically significant. Seventy per cent of retained lower third molars reported in the study were fully erupted. Sixty per cent were vertically placed. There was an association between age and number of teeth present (chi 2 = 38.85, 4DF, P < 0.05), older patients having fewer lower teeth. These data suggest that a large number of patients can expect to keep their lower third molars beyond the age of 35 years and that in many cases a conservative "wait and see' policy for lower third molars in the early twenties is appropriate.

Performance of a computer simulated neural network trained to categorise normal, premalignant and malignant oral smears

The accurate detection of malignant neoplasms whilst they are still small is recognised as one of the main factors increasing chances of survival. Neural networks have many biomedical applications and they have been applied to neoplasia but their use in oral pathology has only recently been documented. The objectives of this study were to train networks to discriminate between normal and dysplastic mucosa. Each network was trained by back propagation, internal cross validation and tested on additional data. The data were derived by analysing 348 intra-oral smears and included mean nuclear and mean cytoplasmic areas of the smears measured by image analysis. A neural network differentiated between normal/non-dysplastic mucosa and dysplastic/malignant mucosa (specificity 0.82, sensitivity 0.76). These early results suggest that integrating neural networks and image analysis, as well as investigating additional criteria, could enhance automation and accuracy of smear techniques in diagnosing oral malignancy.

A comparison of the ability of intra-oral and extra-oral fibroblasts to stimulate extracellular matrix reorganization in a model of wound contraction

Intra-oral wounds, like wounds in children, demonstrate privileged healing when compared with adult wounds at extra-oral sites. This study investigated whether this preferential healing is related to an increased ability of oral mucosal fibroblasts to reorganize extracellular matrix (ECM) when compared with their dermal counterparts. ECM reorganization was investigated by means of a fibroblast-populated collagen lattice (FPCL) system. The effect of donor age was also investigated in this system. Differences in ECM reorganization and FPCL contraction were evident: FPCL contraction was more rapid by oral mucosal fibroblasts than dermal fibroblasts (p < 0.01). FPCL contraction was also greater in child (donor < 10 years) than adult (donor > 18 years) oral mucosal fibroblasts (p < 0.01). These differences were not related to phenotypic differences in cell viability (p > 0.5), DNA synthesis (p > 0.05), and cell number (p > 0.5) within the FPCLs, or cellular attachment to collagen (p > 0.07). FPCL contraction was not stimulated by the addition of conditioned medium from oral mucosal or dermal fibroblasts (p > 0.05). These data show that the significantly increased ability of oral mucosal fibroblasts to reorganize ECM in vitro, when compared with dermal fibroblasts, represents a distinct phenotypic contractile difference, rather than differences in their production of soluble mediators or cell attachment to ECM.