Statistics from Altmetric.com
Arch Dis Child 2001 Volume 84 No 5
H pylori—which antibiotic?
Medical students learn that before prescribing an antibiotic it's a good idea to know which one the offending organism doesn't like. If the facilities are available, resistance and sensitivity testing provides a gold standard. Yet the regimes advised for treatingHelicobacter pylori do not mention such sophistication. Street and others from Parma correct this omission in this month's ADC (page 419). They treated 75 H pylori infected children with a standard antibiotic combination and the next 75 with a combination based on the results of susceptibility tests. Six months later, the authors tested for eradication by endoscopic biopsy histology and rapid urease. When treated “blindly” the organism was eradicated in 83%, rising to 98% after treatment based on susceptibility. The one failure in the latter group was a patient receiving corticosteroids for Crohn's disease.
The authors point out that the cost of testing is offset by savings from fewer repeat treatment courses and—they hope—a lesser risk of resistant strains emerging.
Now all we have to do is decide which children need treating at all.
Little bellyachers (part 513)
In 1995, we published a paper describing a controlled trial of pizotifen in abdominal migraine.1-1 The results were so convincing that the authors reported that they had brought the trial to a premature close because it would have been unethical to continue with the placebo arm. There was lively argument around the editorial table as to the acceptability of a trial involving only 14 children, resolved by the conviction of our statistical adviser. Subsequently the paper was shortlisted to be reviewed in Evidence Based Medicine, so editors, authors, and statistician were vindicated.
The question that this finding raised was how clinicians should differentiate abdominal migraine from other forms of recurrent abdominal pain. Indeed some—possibly many—were unconvinced that such a task was feasible. This month, one of the authors of the 1995 paper, and colleagues from Aberdeen revisit the problem (page 415). They followed 54 children who obeyed the authors' criteria for diagnosis (listed in the paper) 7–10 years after the diagnosis had been made. Nearly 40% still suffered bouts of abdominal pain while just over 70% were current or previous sufferers from migraine headaches.
The authors are at pains to point out that their diagnostic criteria exclude 94% of children with abdominal pain and the outcomes they now report add credence to the robustness of their diagnosis.
First, an audit
Three papers this month reflect the interface between research and audit. The British Society for Paediatric Endocrinology and Diabetes Clinical Trials/Audit Group (not much chance of a snappy acronym) has identified, so far as it is possible to do so, all children in the UK receiving growth hormone on 1 October 1998 (page 387). The background was the change from prescription approval by a national committee (until 1985) to prescribing by individual paediatric endocrinologists and paediatricians. One concern was how closely they were sticking to licensed indications. The answer turns out to be better than most2-2 with only 22% of prescriptions for the 2395 children being “off licence”. The numbers treated for growth hormone deficiency are fewer than reports from the MRC in 1979 and Utah in 1994 despite more oncology patients with iatrogenic GH deficiency. Intriguingly the authors speculate that prevalence may be declining because of the fall in non-surgical breech delivery.
Next, an auditable protocol
As we have said before, we prefer others to publish guidelines. Nevertheless no rule should ever go completely unbroken (even the one about split infinitives). Where we perceive anxiety or uncertaintyADC seeks to soothe and inform. This month Hill & Taylor provide their protocol for treating attention deficit/hyperactivity disorder (ADHD) (page 404). What made the commissioning editor regard their contribution as special is the fact that all is auditable. We hope that departments who adopt the protocol will take the point. Please note that a complete reading of the paper demands logging on to the paper's electronic version onwww.archdischild.com where the necessary questionnaires are lodged.
Finally we need an audit
Several acute illnesses are magnets for the media. In the UK, clinicians have had to contend with press hysteria over flesh eating killer bugs, autism, meningitis (in reality meningococcal septicaemia) and expiring in a fast food restaurant courtesy of a nut-contaminated burger. Parents pressure general practitioners, and GPs, albeit probably more gently, pressure paediatricians into providing an epinephrine (adrenaline) syringe for children who have experienced an unpleasant experience attributed to food allergy. If you look for the evidence behind the trend it is vanishingly small.
We invited Dr Unsworth, from Bristol (the town not the pharmaceutical manufacturer) to put the problem in perspective (page 410). He warns us of the potential hazards of this particular current enthusiasm.
Let's hope this won't be yet another example where we will weakly tell our successors that it seemed like a good idea at the time.