Inherited metabolic diseases (IMDs) pose a particular challenge to diagnosis. Although individually rare, improved diagnostics and greater awareness have shown that the incidence is much greater than previously thought. The commonest disorders such as phenylketonuria and medium chain acyl-coA dehydrogenase deficiency (MCADD) have an incidence of approximately 1 in 10 000; however, collectively, all IMDs have an incidence of <1 in 1000.1 It is important to make IMD diagnoses because there are many effective treatments and early diagnosis greatly enhances the chance of a better outcome. The individual rarity of the IMDs means that exposure remains limited for most paediatricians unless working in a centre specialising in these disorders, and so experience and confidence in dealing with such patients may not be well developed.

Presentations are also non-specific further increasing the chances of a diagnosis being missed. At present, a sick neonate is more likely to have multiple septic screens undertaken than a metabolic one. Infection is far commoner than an IMD presentation, but it is important to entertain the possibility of an IMD early, allowing the opportunity for effective intervention. The other difficulty is that most IMDs will not be diagnosed unless specific investigations for that disorder are undertaken. An infant with maple syrup urine disease, although very sick, will not be revealed by standard investigations. Plasma amino acids are required to reveal the diagnostic elevation of the branched chain amino acids, leucine, isoleucine and valine. Similarly, the workup of an encephalopathic child should include measurement of ammonia because this is unlikely to be picked up in any other way. It is therefore essential to develop an approach to thinking about and investigating IMDs, particularly when one considers that these patients will present initially to the local hospital rather than to the specialist centre.