Article Text

This article has a correction. Please see:

Sweat sodium is not sweat chloride
  1. S K HALL,


  4. A GREEN
  1. Clinical Chemistry Department and Department of Respiratory Medicine and Cystic Fibrosis*
  2. Birmingham Children’s Hospital
  3. Ladywood Middleway
  4. Ladywood, Birmingham B16 8ET

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Editor,—We read the annotation on the diagnosis of cystic fibrosis with great interest.1 However, some points about the sweat test that were made are potentially misleading and unfairly denigrate a valuable diagnostic tool.

    We were interested in the statement that ‘10% of normal adolescents will have sweat salt concentrations greater than 60 mmol/l’. Although this may be the case for sweat sodium, we do not find this for sweat chloride concentrations, provided a cut off of greater than 70 mmol/l as suggested by the author is used rather than 60 mmol/l as quoted in the seventh paragraph. Sweat sodium concentrations do have a greater age dependency2 and may be misleading or confusing if chloride is not measured. In our experience, interpretation of sweat electrolytes does not require particular ‘correction for age’ as the cut off for sweat chloride of >70 mmol/l that we use for neonates and infancy, is maintained for adults,3 providing excellent discrimination. Reviewing sweat tests performed here on 68 individuals with clinical suspicion of cystic fibrosis aged from 10 to over 40 years over the last five years, seven had clear increases of sodium and chloride consistent with cystic fibrosis. Of the remaining 61, 14 had a sweat sodium greater than 60 mmol/l (with the chloride varying from 36–69 mmol/l) but none had a chloride greater than 70 mmol/l. To our knowledge only one of these 61 patients has subsequently been classified clinically as having cystic fibrosis and was associated with a ‘mild’ genotype. We have found the ratio of chloride to sodium to be a valuable interpretative tool which is demonstrated best in graphical format. When the ratio is used in conjunction with the chloride concentration, patients with and without cystic fibrosis very rarely overlap in any age group in our hands. The fludrocortisone suppression test was used in the investigation of adults in the days when sweat sodium was often measured in isolation and should now be considered unnecessary and out of date for all ages.

    Repeating a sweat test to confirm a positive result is prudent, but performing it three times to avoid technical errors should not be necessary. Centres needing to do this must consider whether they should continue to offer the test at all.

    Difficulties because of inadequate sweat should not arise in infancy, although inadequate sweat weights may be a little more commonplace in neonates.

    In contrast to the author, we consider the sweat test to be very useful indeed in infancy in the diagnosis of cystic fibrosis. It is very unusual in our population in the West Midlands to come across normal sweat chlorides in children who are subsequently found to have cystic fibrosis. In the 1990s it is inappropriate to measure only sweat sodium. Moreover, sweat sodium and sweat chloride are not synonymous and must be differentiated—to discuss sweat ‘salt’ muddies the waters. We believe that the sweat test continues to provide valuable diagnostic information when interpreted correctly in the context of the clinical presentation—don’t knock it out of the diagnostic repertoire.

    Dr Wallis comments:

    I am sorry that your correspondents feel that the sweat test got a bad press in the recent annotation. This was never intended and statements such as ‘the sweat test remains the gold standard for the diagnosis of cystic fibrosis’ and ‘raised sweat electrolytes confirm most cases of cystic fibrosis’ are strong confirmation of its importance. Far from knocking it I can only reconfirm its central place by again quoting from the article’s ‘golden rule’ to perform ‘the sweat test in a centre that undertakes the test regularly and measures both the sodium and chloride’.1-1

    Not all biochemistry departments measure both chloride and sodium levels in sweat routinely. Some believe that sodium levels have been well studied and provide robust information for varying sweat weights in childhood1-4 while others prefer to measure chloride. Most clinicians would like to have levels for both electrolytes in unusual clinical cases or after equivocal results in the initial testing. In these borderline cases—as with the small infant—I would have no problem in requesting a third sweat test.

    I thank the correspondents for emphasising the importance and benefits of measuring and differentiating between both salts in the sweat test and for highlighting the specific value of chloride levels. But a normal sweat chloride, although rare, is well described in cystic fibrosis. Beware the unusual phenotypes. They are out there.


    1. 1-1.
    2. 1-2.
    3. 1-3.
    4. 1-4.