METHODS

The study was undertaken after approval from the relevant local research ethics committees. Sequential paediatric attendances presenting to three urban emergency departments in the North West of England conurbation, after involvement as a passenger in a car crash were prospectively studied, over an eight month period from May 1999 to January 2000.

Children aged 4–16 accompanied by a responsible carer were approached and asked to participate in the study. Consent was gained from the responsible adult and assent from the child. Children requiring resuscitation or inpatient admission for their injuries were excluded.

The responsible parent or adult was contacted at 48 hours after discharge from the emergency department and a structured interview was performed, by telephone, using a structured questionnaire. A single researcher (RB) made telephone contact using a standardised questionnaire using a “template” sequence of closed questions to elicit “Yes/No” answers. These questions were directed to elicit the presence or absence of any of the following—complaint of neck pain, objective evidence of neck discomfort, reduced range of motion in the neck/upper trunk, time taken off from school, or evidence of restricted physical activity. The order and wording of each telephone interview was identical. If symptoms of neck pain, discomfort, or decreased motion were reported on direct questioning an appointment for formal clinical review was made for approximately five days after the initial car crash. If no symptoms of pain, discomfort or decreased motion were elicited on direct questioning then, in accordance with the Quebec Taskforce criteria, a grade of WAD grade 0 was given.

Clinical review of symptomatic children was performed at day 5 after the car crash by a single researcher (RB) to avoid interobserver variability. A structured clinical examination of the cervical spine (occiput to the clinical cervicothoracic junction) and history was taken as per the minimum dataset suggested by the Quebec Task Force group.⁴ A clinical grading of the WAD was also made using the Quebec Task Force criteria (table 1). Further clinical review for those with WAD of grade 1 or greater was then arranged for 14 days, 28 days, and 56 days. Patients were discharged from follow up when they became asymptomatic, namely, WAD grade 0. A further telephone follow up was then performed after clinical discharge to confirm the child's continued asymptomatic state. Follow up was completed for all children included in the study. Chart reviews were undertaken to ascertain the use of radiological investigation and the formal radiologist report for those investigations was obtained. Radiographs were ordered in accordance with existing protocols at each site.

Data were analysed using a SPSS statistical package for PCs using the χ² test, the Mann-Whitney U test, and Kaplan-Meier survival graphs.

RESULTS

Altogether 110 sequential attendances were identified and included in the study over an eight month period. Five attendances were excluded as consent or assent was not given.

Thirty nine per cent of children studied were front seat passengers, with 61% as rear seat passengers. Ninety one per cent were reported as being suitably restrained in either, 3 point/4 point or “lap strap” restraints. Thirty two per cent were involved in frontal impacts, 18% in side impacts, and 50% in rear impact car crashes.

The overall incidence of WAD was 47% (49 of 105). Sixty per cent (29 of 49) of the children with WAD were symptomatic on the day of presentation to the emergency department with the remaining 40% of children with WAD becoming symptomatic the following day.

Fifty five per cent (27 of 49) of the WAD affected children were girls. There was no statistical difference in terms of incidence of WAD between the male and female groups on χ² testing.

Children who developed WAD symptoms were distributed over the age range studied as shown in fig 1. There was a non-significant trend to an increased number of WAD symptomatic patients in the older age groups.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Age distribution of children with WAD.

Radiological investigation was performed in nine children of which four had WAD grade 1 and five had WAD grade 2. There was a significant difference using χ² testing in terms of the use of radiology between the children with WAD grades 1 and 2 (p<0.05). All cervical spine radiographs were reported as showing no abnormality by consultant radiology staff.

Clinical examination of the WAD group demonstrated nine children with Quebec Task Force WAD of grade 2 and 40 children with a WAD of grade 1. The overall mean duration of symptoms for children with WAD was 8.8 days (range 2–62). The mean duration of symptoms for the WAD grade 2 children was 19.7days (range 2–62). The mean duration of symptoms for the WAD grade 1 children was 6.4 days (range 2–18). There was a significant difference in duration of symptoms between the WAD grade 1 and 2 groups on testing with the Mann-Whitney U test (p=0.034). Figure 2 is a Kaplan Meier survival graph showing the resolution of symptoms against time for the WAD grade 1 and grade 2 groups. No children reached the accepted criteria for chronicity of WAD namely pain lasting more than two months, as defined by the Quebec Task Force group.⁴

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Clinical course of WAD grades 1 and 2 in children.

DISCUSSION

WAD has until now almost exclusively been reported as an adult phenomenon with very few children identified in previous cohorts. This study is the largest clinically assessed group of children with WAD to date. Recent work from the UK ⁵ had alluded to an appreciable incidence of WAD in children but had not been able to clinically confirm this. Hadfield et al suggested an incidence of 29.5%, a figure somewhat lower than that oft quoted for the incidence of WAD in adults. Our figure of 47% is appreciably higher than many previous adult emergency department reports and also higher than previous insurance company data.⁶ One obvious factor responsible for this may be the “positive” reporting bias inherent in patients attending emergency departments. Measures were taken, as far as practicable, to reduce the use of leading questions in the initial structured telephone questionnaire to avoid an over-reporting bias. Such high rates of WAD were unexpected given the paucity of previously published evidence and does suggest a hidden incidence that may not be being recorded in children presenting after car crashes.

The well known time delay before the onset of symptoms in adults⁸ was reflected also in our paediatric population. Forty per cent of children with WAD experienced the onset of symptoms over 24 hours after the car crash. As before it is not fully understood why such a delay in onset of symptoms should occur but may reflect the soft tissue injury aspect of WAD with late muscle stiffness and pain.

Cervical spine radiographs are taken infrequently in children. In our series only nine cervical spine series were requested in children with WAD grade 1 and 2. Radiographs were ordered according to local practice guidelines. Of note is the increased rate of use in WAD grade 2 but that being so, not all this group did in fact have radiographs performed. Applying the Quebec Task Force guidelines for radiography in WAD⁴ would significantly increase the number of radiographs performed in children as it suggests that all patients with a WAD grade 2 or above should have cervical spine radiographs performed. Further work is necessary to evaluate if this is in fact beneficial in children.

The clinical course for WAD in children would seem to be much more favourable than that in adults. The Quebec group suggested a mean time to resolution of 28 days. Our cohort had a mean time to resolution of 8.8 days. Intuitively one would have expected that the more severe WAD, grade 2 would take longer to resolve and this was borne out in our study. The reasons for the much more favourable outcome are probably multifactorial and almost certainly include both physical and psychological factors. Recent work has shown that children are susceptible to and do experience significant psychological trauma after being involved in a car crash.⁹ This study did not attempt to differentiate the physical from psychological components of the reported symptoms after the car crash, more quality research is needed to elucidate this.

Contributors

RB and DWY initiated and designed the study. RM, RB and LD prepared the patient and guardian information and participated in the ethics committees stage. RB collected the data with telephone and clinical interviews. RB analysed the data and wrote the paper with DWY. All authors have contributed to the drafting of the submitted paper. RB acts as guarantor for the paper.