Diagnosis after a first paroxysmal event may be difficult; even experienced clinicians may disagree. In the Netherlands (Journal of Neurology, Neurosurgery, and Psychiatry2004;75:241–5OpenUrlAbstract/FREE Full Text) six experienced paediatric neurologists (two panels of three each) were presented with detailed descriptions of 100 first paroxysmal events in children. Using clinical judgement the individual paediatric neurologists reached only 41% agreement about seizure diagnosis (kappa (degree of agreement beyond chance) = 0.41). When they used defined criteria the degree of agreement was 45%. The two panels were able to reach 60% agreement after discussion within each panel. For seizure classification agreement between individuals was 46% using clinical judgement and 57% using defined criteria. Agreement between panels was 69%. The panels could agree on syndrome diagnosis for only 24 of 51 children considered to have had a first seizure.

Congenital leukaemia is rare, not always fatal, and to be distinguished from transient myeloproliferative disorder (TMD). In the Northern Health Region of England during an 18-year period (October 1984–September 2002) nine infants were referred to the central unit with a diagnosis of possible acute leukaemia arising before the age of 3 months (Journal of Clinical Pathology2004;57:186–8)OpenUrlAbstract/FREE Full Text. Five had acute leukaemia and four TMD. Of the four infants with TMD three had Down’s syndrome and the fourth had trisomy 21 in proliferating marrow cells. The incidence of acute leukaemia was calculated to be 8.6 per million live births per year. Three of the five infants with acute leukaemia had remission with treatment but only one has survived and she remains well at the age of 5 years after presenting at birth. One child with Down’s syndrome died at the age of 1 day with acute myeloid leukaemia. All four children with TMD have survived but one, with Down’s syndrome, later developed acute myeloid leukaemia.

Methotrexate is effective treatment for juvenile idiopathic arthritis (JIA) but stopping the treatment results in reactivation of disease in about half of patients. Now researchers in Germany (Annals of the Rheumatic Diseases2004;63:206–8OpenUrlAbstract/FREE Full Text) have suggested that measurement of serum concentrations of myeloid related proteins 8 and 14 (MRP8/MRP14) might help to decide when to stop methotrexate treatment. Of 22 patients with JIA in remission on methotrexate 15 had serum MRP8/MRP14 concentrations at or above 250 ng/ml. Twelve of these, but none of the seven with lower concentrations, relapsed within 1 year of stopping methotrexate. A serum MRP8/MRP14 concentration of 250 ng/ml or higher was 100% sensitive and 70% specific for relapse within 1 year.

A survey of 13 Welsh hospitals (Journal of Clinical Pathology2004;57:126–30OpenUrlAbstract/FREE Full Text) has revealed a gallimaufry of practices in the investigation and management of short stature. Five hospitals screened fewer than five children a year and in some hospitals an appreciable proportion of children (up to 70%) with a diagnosis of growth hormone deficiency were not treated with growth hormone. Clinical assessment and investigations varied between hospitals. For one test (clonidine stimulation) eight different protocols were in use. In the interpretation of growth hormone stimulation tests nine different cut-off values for defining abnormal responses were quoted and in three hospitals paediatricians and biochemists were using different cut-off values. Guidelines based on published opinion are presented with the paper.

If early-life bacterial infections reduce the likelihood of later asthma then giving antibiotics to children might make later asthma more likely. A retrospective cohort study of 746 adults at three general practices in Kent (Thorax2004;59:11–15OpenUrlAbstract/FREE Full Text) has led to the conclusion that the association between childhood antibiotic use and later asthma is most likely an example of reverse causation; young children with asthma may be given antibiotics for their chest symptoms. The association is strongest for asthma and weakest for atopy (positive skin tests) and applies to antibiotics prescribed for chest symptoms but not for other indications.

Highly active antiretroviral therapy (HAART) including two nucleoside reverse transcriptase inhibitors and either a protease inhibitor or a non-nucleoside reverse transcriptase inhibitor, is effective treatment for both adults and children with HIV infection but the best time to start treatment in children is uncertain. Data from the USA (Lancet2003;362:2045–51OpenUrlCrossRefPubMedWeb of Science) show that protease inhibitor based HAART raises the percentage of CD4+ cells in HIV-infected children but is most effective when treatment is started at younger ages and before severe immunosuppression has occurred. Randomised trials are needed to establish the best time to start treatment.

The prevalence of vitamin A deficiency in a population is assessed by measuring plasma retinol concentrations. These concentrations are, however, lowered by clinical or subclinical infection. A meta-analysis of published data (Lancet2003;362:2052–8OpenUrlCrossRefPubMedWeb of Science) has shown that increases in acute phase proteins (C-reactive protein, α1-acid-glycoprotein, α1-antichymotrypsin, and serum amyloid A were considered) are associated with decreased concentrations of retinol in plasma. It is recommended that in population surveys C-reactive protein and α1-acid-glycoprotein concentration should be measured along with plasma retinol concentrations. Appropriate correction of retinol concentrations would give a truer estimate of the prevalence of vitamin A deficiency.

In patients with cystic fibrosis the CFTR mutation determines the degree of pancreatic damage but the association between CFTR mutation and lung damage is much less clear. Modifier genes may be involved and the nitric oxide synthase 1 gene (NOS1) is regarded as a prime candidate. Workers in Paris (Thorax2004;59:156–8OpenUrlAbstract/FREE Full Text) have shown an association between NOS1 genotype and decline in lung function. The number of GT repeats in the 5-untranslated region of this gene is very variable; it varied from 18 to 36 in 118 subjects (59 adults with cystic fibrosis and 59 controls). In both patients and controls a greater number of repeats was associated with an increase in concentration of nitric oxide in expired air. In patients a greater number of repeats was associated with slower decline in lung function (annual percentage decline in FEV₁ 3.3%, 3.2%, and 0.8% in patients with 0, 1, and 2 NOS1 alleles with >27 repeats). The findings were unaffected by age, bacterial colonisation, or CFTR genotype.

Lucina is fascinated by daft names but she’s come across none dafter than this one. Dickkopf 1, or DKK1 (there must be more than one thickhead protein) is a soluble inhibitor protein that acts on the Wnt family of growth factors and is important in skeletal development. Defective binding of DKK1 to a Wnt coreceptor causes hereditary high bone density and manipulation of DKK1 concentrations in embryonic animals leads to skeletal abnormalities. Researchers in Arkansas (New England Journal of Medicine2003;349:2483–94OpenUrlCrossRefPubMedWeb of Science) have shown that overproduction of DKK1 by myeloma cells is associated with the development of lytic bone lesions in adults with multiple myeloma. It inhibits the development of osteoblasts.