Piscator's cup overfloweth on this rare occasion – how can one journal manage to come up trumps in respect of articles that excite and inform in equal measure? OK the long, dark nights may explain why this would pass for excitement in this very northern latitude but my heart skipped a beat mid-winter on spotting at least eight articles in last year's Journal of Pediatrics with something to say about that old chestnut – cystic fibrosis (CF).

Screening for CF has probably now rolled out nationwide in the UK and the laboratory's involvement has been detailed in the commendable national guidelines for sweat testing, which are in the process of revision. That screening has thrown up some interesting results cannot be understated and, as a consequence, the definition of normal has come up for scrutiny. Under a strapline of ‘the sweat of our brows’, an editorial (J Pediatr 2008;153:735–6) comments on two articles in the edition that re-examine reference ranges for sweat electrolytes. As the editorial itself notes, the information both clarifies and confuses so here's hoping the UK Guideline group have more success. The first article reports sweat chloride (and sodium) in 40 individuals in each of six age categories (J Pediatr 2008;153:758–63). In assembling the reference population, the authors excluded individuals with chronic respiratory symptoms, thereby minimizing the likelihood of inadvertently recruiting subjects with mild CF, carriers of ΔF508 mutations and relatives of known patients with CF. The data corroborated the traditional cut-off of less than 40 mmol/L for normal sweat chloride in childhood, but many unaffected adults had sweat chloride concentrations in the range considered equivocal, 40–60 mmol/L. This information is important as increasingly adults with chronic respiratory, reproductive and pancreatic symptoms are having sweat tests. Conversely, the authors of another paper in the same edition (J Pediatr 2008;153:857–9) argue that the lower limit of normal for neonates screened for CF and found to have at least one known mutation should be lowered from 30 to 24 mmol/L. These data fly in the face of data already accumulated for newborns identified as heterozygote carriers and raise issues as to what constitutes disease in genetic variants. Prima non nocere

On this side of the Atlantic and in a formidable show of cooperative endeavour, the United States of Europe have assembled a consensus document on what exactly to do with equivocal test results from newborn screening for CF (J Cyst Fibros 2009;8:71–8). The authors endorse the use of 30–60 mmol/L as the equivocal range for sweat chloride in follow-up testing of newborns. On the other side of the Atlantic, the Cystic Fibrosis Foundation has published guidelines for the diagnosis of CF throughout life (J Pediatr 2008;153:S4–14). They endorse the equivocal range adopted by their European counterparts but advise that as more data emerge from the Newborn Screening programmes the lower limit of the equivocal range will have to be reduced – something that the editorial referenced above disputes. This document is an update of their earlier guidance on diagnostics published over a decade earlier. In the intervening years, the number of mutations has increased to 1500 and the ability to sequence the whole gene for diagnostic testing is routinely available. There is a greater understanding of genotype–phenotype relationships, which is helping to establish a break point for what constitutes a clinical diagnosis of CF particularly in patients presenting later in life with symptoms as diverse as chronic sinusitis and infertility.

Our own journal has also carried a paper on the analysis of sweat chloride in young infants (Ann Clin Biochem 2009;46:73–8) in which sweat collected onto filter paper (from which chloride was subsequently eluted and measured by coulometry) was compared with chloride measured in liquid sweat analysed by the ‘additions’ method using ion-selective electrodes. This last study validated the use of the ‘additions’ method for measurement of sweat chloride in liquid sweat, although admittedly the bias typically observed between the two methods for measurement of sweat chloride was reversed with liquid sweat chloride demonstrating a positive bias compared with chloride eluted from the collection paper.

Finally, sweat testing is yet another unsuspected opportunity for parental interference.

A salutary case report (Clin Biochem 2008;41:1110–2) describes the usefulness of measuring sweat potassium as a means of establishing that the mother of a two-year-old boy was interfering with the collection of sweat. A presumptive diagnosis of CF in the child, which prompted numerous sweat tests and exhaustive genetic testing, was revised in favour of one of Munchausen by proxy.