Dear Editor,

We read with interest the study of Dunning et al [1]. We have done a similar audit in our department and would like to share our results.

As in the study, our own local guidelines on the management of children presenting with head injury, are based on the guidelines of the Royal College of Surgeons (RCS) of England published in June 1999 [2]. The aim of our audit was to ensure that children with head injury are receiving appropriate management in line with the local protocol and to compare local practice against new recommendations recently published by the National Institute of Clinical Excellence (NICE) [3].

Methods: We retrospectively collected data on all children under 16 years of age, presenting to the Accident and Emergency department of Scunthorpe General Hospital, with a head injury of any severity, during the period 1st July 2003 to 31st October 2003. Our outcome measures were admission criteria, Skull radiograph (SXR), brain computed tomography (CT) and discharge criteria. Actual management was compared against the RCS and NICE guidelines

Results: A total of 361 children presented with a head injury within the 4-month period (Table 1).

a) Admission criteria: 29 out of the 361 children were admitted (8.0%). According to the RCS guidelines, only 25 children should have been admitted (6.9%). However we also admitted 4 children with minor head injury. According to NICE guidelines admission is intrinsically linked to CT results. The admissions will be much lower and we predict we would have admitted 11 (3.0%) children.

b) Skull radiograph: Of the 361 children, SXR was performed in 60 children (16.6%). As per the RCS guidelines SXR should have been performed in 25.2% children. If we had used NICE guidelines only 2 children would have needed SXR (0.6%).

c) CT Scan: A total of 3 CT scans were done (0.8%). In this regard RCS guidelines were followed in 100% cases. If we strictly apply NICE guidelines CT would have been indicated in 20 children (5.2%).

d) Discharge criteria: 26 children had normal observations and were discharged within 24 hours. One child was transferred to a tertiary neuro- surgical centre as he had a depressed skull fracture. He needed a CT scan, requiring Paediatric anaesthesia, which is not available in our institution. One child was kept for more than 24 hours due to concerns about non-accidental injury and another for fractured mandible. 91% of discharged children were given an advice card.

Discussion: If NICE guidelines are strictly followed in a district general hospital like ours the admission rates for head injury are likely to fall from 8.0% to 3.0 %. There will also be fewer skull radiographs (16.6% compared to 0.6%), however the CT scan rate will increase from 0.8% to 5.2%. NICE guidelines state these scans should be performed and reported within one hour. For our hospital this would mean 17 more scans in 4 months or approximately 51 scans per year (in children less than 16) if we have a similar case load throughout the year. A number of these scans would have to be done after 5 p.m. This will have severe unfunded resource implications and currently radiology services are unable to cope with this extra demand.

Small children need to be sedated for CT scan (11 out of the 29 [37.9%] children admitted were less than 5 years old). In the very young ones this means they will need a paediatric anaesthetist. Most district general hospitals do not have paediatric anaesthetists available round the clock. Therefore these patients would have to be transferred to tertiary centres (One of our patients' had to be transferred for this reason). This will increase the load on tertiary centres.

One of the indications for a CT scan in the NICE guidelines is Glasgow coma scale equal to 13 or 14 two hours after the injury. Any crying young child seen 2 hours after injury would score 14 on the Glasgow coma scale and would therefore need a CT scan. This would lead to a lot of probably unnecessary scans in young children.

Our admission rates are higher than those reported by Dunning et al (8.0% Vs 3.7%). However there were 511 children (4.7%) in their study who were observed in the emergency department for more than 3 hours. In our hospital most of these children would have been observed on the Paediatric ward and would have been counted as an admission. Therefore if we add these 511 children, then our admission figures are similar.

In our series SXR was performed in only 16.6% children compared to 25% in the study by Dunning et al. There have been questions about the utility of SXR in children with head injury (4) and this may have been the reason why doctors in our hospital ordered less SXR.

In conclusion, our results are quite similar to the study by Dunning et al and shows the extra burden that would be placed on radiology departments if the NICE guidelines are fully implemented.

References

[1]. Dunning J, Daly JP, Malhotra R, et al. The implications of NICE guidelines on the management of children presenting with head injury. Arch Dis Child 2004;89:763-767.

[2]. Royal College of Surgeons of England. A working party on the management of the head injured patient. Holborn, London: RCS England, Lincoln’s Inn Fields, 1999.

[3]. The National Collaborative Centre for Acute Care. Head injury: triage, assessment, investigation and early management of head injury in infants, children and adults. Clinical Guideline N0234. Published by National Institute of Clinical Excellence, http://www.nice.org.uk 2003.

[4]. Boulis ZF, Dick R, Barnes NR. Head injuries in children—aetiology, symptoms, physical findings and x-ray wastage. Br J Radiol 1978;51:851–4.

Table 1. Management of 362 children with head injury

Dear Editor,

We are concerned that the conclusions of the recent article by Christy et al [1] are not supported by their data, and that their article might be misinterpreted by some as casting doubt on the usefulness of influenza vaccine among asthmatics.

In observational studies such as the one by Christy, a crucial issue in interpreting the results is understanding how the authors went about controlling for the underlying reasons that would determine, among asthmatics, why certain of them received influenza vaccine and why others did not. As one example, patients with severe asthma could have been more regular patients of the clinic and were thus more likely to be vaccinated and have exacerbations detected there. Sporadic users of clinic services would have been less likely to have been vaccinated there (if at all) and may not have had their exacerbations treated there. Dichotomizing utilization, as Christy did, or classifying asthma as mild, moderate or severe, may not have completely controlled for this behavior in the analysis.

In a prior study done by us, we found that controlling extensively for identifiable ‘characteristics’ of severity among the asthmatic subjects, such as number of prior emergency department visits or utilization of asthma medication provided only incomplete control of the extent of the severity of the level of asthma. [2] In that past study, we used unconditional Poisson regression models that simultaneously adjusted for severity of asthma and for sex, age, HMO, preventive care practices, and seasonal fluctuations in asthma exacerbations. To control for severity of asthma, we used the number of inhaled beta-agonist dispensings, the number of hospitalizations and ED visits for asthma during the 6 months before the influenza season, and we also adjusted for preventive care practices with frequency of cromolyn dispensings during the 6 months before the influenza season. Even so, in this analysis, the relative risk for asthma exacerbations among those vaccinated ranged from 1.4 to 2.2, similar to that found by Christy et al.

It was only when we analyzed our data using a self-control method of adjustment that influenza vaccine was shown to be efficacious in reducing asthma exacerbations. Using the self-control method, incidence rate ratios of asthma exacerbations after vaccination ranged from 0.59 to 0.78 during 3 influenza seasons, and the protective effect was statistically significant in each season. Christy et al. state that ‘the adverse effect of vaccination disappeared only after the authors eliminated the unvaccinated group from the analysis, but this is a questionable methodology for assessing vaccine effectiveness’. The self control methodology is one that is well accepted, has been used for many studies - including ones of adverse events following immunization, myocardial infarction following strenuous physical activity, and motor vehicle accidents associated with cell phone usage - and allows for more exact adjustment of individual characteristics that would otherwise confound observational studies. [3-6] Our self-control analysis had not simply eliminated the unvaccinated group from the analysis, instead it focused on comparing rates of acute asthma attack before versus after influenza vaccination. The self-control method used each subject’s pre-vaccination rate as a baseline rate unique to that subject, thereby avoiding confounding by differences in asthma severity between subjects. Although most commonly used for events that occur very closely linked in time to some proposed trigger, this method can also be applied to events more distant in time, as we did in our study. [7]

In conclusion, we disagree with the authors’ claim that self-control analyses are not valid. Our previous study suggests that influenza vaccination may protect against asthma exacerbation.

References

(1). Christy C, Aligne CA, Auinger P,et al. Effectiveness of influenza vaccine for the prevention of asthma exacerbations. Arch Dis Child 2004:89:734-735

(2). Kramarz P, Destefano F, Gargiullo PM et al. Does influenza vaccination prevent asthma exacerbations in children. J Pediatr 2001; 138:306-10.

(3). Albert CM, Mittleman MA, Chae CU, et al. Triggering of sudden death from cardiac causes by vigorous exertion. N Engl J Med 2000;343:1355-61.

(4). Farrington CP, Nash J, Miller E. Case series analysis of adverse reactions to vaccines: a comparative evaluation. Am J Epidemiol 1996;143:1165-73.

(5). Mittleman MA, Mintzer D, Maclure M et al. Triggering of myocardial infarction by cocaine Circulation. 1999;99:2737-41

(6). Riddlemeier DA, Tibshirani RJ. Association between cellular- telephone calls and motor vehicle collisions. N Engl J Med. 1997;336:453- 8.

(7). Maclure M, Mittleman MA. Should we use a case-crossover design? Annu Rev Public Health 2000: 21:193-221.

Dear Editor,

We read with interest the recent article by Wright et al[1], which pointed out that most infants are weaned before 4 months of age, and not at 6 months as recommended by the Department of Health.

We addressed similar issues in a questionnaire survey of weaning practice in Wirral, Merseyside. This was conducted in May and June 2004, at the time of the 8-month health visitor appointment. Of the 116 mothers who responded, 15 (12%) chose to wean their infant at 3 months, 72 (59%) at 4 months and 22 (18%) at 5 months. Only 7 respondents weaned their infants at 6 months or later (6 weaned at 6 months and 1 at 7 months). Thus our findings are consistent with those in the Glasgow study.

Respondents were asked why they chose to wean at a particular age and were asked to categorise responses as ‘baby not satisfied with milk’, ‘I was advised to’ or as a free text reply. The commonest reason (97 mothers, 80% of respondents) was that the infant was not satisfied with milk alone, 18 (13%) mothers introduced solids because of external advice and 19 (14%) gave free text answers.

Six respondents weaned due to their infant not sleeping through the night. Three mothers cited experience with other children as influencing their decision. Other reasons were ‘returning to work’, ‘to help with colic’, ‘because it was the right time’ and ‘because I felt he was ready’.

Our findings support those of Wright et al, that parents’ decisions on weaning are based on their perceptions of the needs of their infants rather than on external advice. The findings also confirm that most parents are weaning earlier than the recommended 6 months and pose a challenge for the successful implementation of this recommendation.

Reference

(1). Wright CM, Parkinson KN, Drewett RF. Why are babies weaned early? Data from a prospective population based study. Arch Dis Child 2004; 89: 813-6.

Dear Editor,

In their timely review regarding gender assignment in newborns with physical intersex conditions, Ahmed et al [1] made a cursory reference to the ‘John v Joan case,’ noting that it ‘famously challenged’ John Money’s hypothesis about the importance of the rearing environment vis-á-vis gender identity differentiation and formation. Ahmed et al comment that ‘this single, rather atypical case of gender identity change from female to male gender may have over-exaggerated the role of prenatal hormones in the development of gender identity in all cases of complex genital anomalies...’ (p 848).

Ahmed et al do not really provide the reader with any specific information about the nature of the John v Joan case. The patient, whose real name was David Reimer, was, at birth, a normal biological male. He happened to have an identical twin. At the age of 7 mos, Reimer’s penis was literally burned off during electrocautery for circumcision, which was required due to phimosis. A decision was made to raise Reimer as a girl around the age of 17 mos, and castration and initial genital reconstruction were instituted around the age of 21 mos. [2].

Because Reimer was born a normal biological male, some theorists touted this case as one of the most stringent tests of an ‘experiment of nurture’ with regard to gender identity differentiation. When it was subsequently reported that Reimer appeared to be developing a female gender identity during childhood, it was argued that this was evidence in support of the rearing environment in determining gender identity formation [3]. Unless Reimer was ‘predisposed’ to be a male-to-female transsexual, how else could one account for a normal biological male apparently developing a female gender identity?

Subsequent follow-up reports of Reimer in adolescence and adulthood indicated that the ‘experiment of nurture’ did not work. By early adolescence, Reimer had rejected his identity as a female, ‘reverted’ to the male social role, began to receive testosterone injections, and efforts were made at phallic reconstruction. Reimer’s sexual orientation appeared to be exclusively sexually attracted to women. At age 25, he married a woman and adopted her children. As a result, the long-term outcome of the Reimer case has been used by other theorists to challenge the importance of the rearing environment in determining gender identity formation [4]. Moreover, many critics have used the case to challenge the general principle that the rearing environment is an important factor in gender identity formation and, as a result, clinicians are now more uncertain about making recommendations for gender assignment with regard to the various physical intersex conditions that were reviewed by Ahmed et al. (As an aside, two other tragic developments about the Reimer case are worth noting. In 2002, Reimer’s twin brother committed suicide and, in 2004, at the age of 38, Reimer also committed suicide[5].)

In their review, Ahmed et al failed to mention another case of ablatio penis that our group described [6]. Our patient’s penis was also burned off during electrocautery, at the age of 2 mos. A decision to raise our patient as a girl was made at 7 mos, at which time castration was performed. At the age of 26 years, our patient was interviewed for a psychosexual follow-up. Our patient reported a female gender identity. Her recalled gender role behaviour in childhood was ‘tomboyish’. As an adult, she worked in an occupation typically populated by men. Her sexual orientation in both phantasy and behaviour was bisexual.

The gender identity outcome of our patient was, therefore, strikingly different than in the Reimer case. Both Reimer and our patient were, however, ‘tomboyish’ during childhood. Whereas Reimer appeared to be exclusively attracted sexually to women, our patient was sexually attracted to both women and men. Thus, there were both similarities and differences in psychosexual differentiation in these two cases of ablatio penis.

We agree with Ahmed et al that the outcome in the Reimer case should be interpreted cautiously with regard to clinical management principles for patients born with physical intersex conditions. The outcome of our own case makes the matter even more complicated.

References

(1). Ahmed SF, Morrison I, Hughes IA. Intersex and gender assignment; the third way? Arch Dis Child 2004;89:847-50

(2). Zucker KJ. Intersexuality and gender identity differentiation. Annu Rev Sex Res 1999;10:1-69

(3). Money J. Ablatio penis: normal male infant sex-reassigned as a girl. Arch Sex Behav 1975;4:65-71

(4). Colapinto J. As nature made him: the boy who was raised as a girl. Toronto: HarperCollins, 2000

(5). Chalmers A. Boy raised as girl commits suicide. The National Post, May 10, 2004

(6). Bradley SJ, Oliver GD, Chernick AB, Zucker KJ. Experiment of nurture: ablatio penis at 2 months, sex reassignment at 7 months, and a psychosexual follow-up in young adulthood. Pediatrics 1998;102:E91-E95. (http://www.pediatrics.org/cgi/content/full/102/1/e9)