Sweat testing for cystic fibrosis: How good is your laboratory?

Most paediatricians are aware of the fundamental role that the sweat test has played in confirming or excluding the diagnosis of cystic fibrosis (CF).1 However, many would be unaware of the technical aspects of performing, analysing and interpreting the sweat test, and most blissfully unaware of how their sweat test laboratory matches up to accepted guidelines.2–4 In this issue of the Journal, Mackay et al. report the disturbing results of a questionnaire survey of clinical laboratories offering sweat tests in New Zealand.5 The only similar report from Australia was from New South Wales in the context of newborn screening for CF and also reported a disturbing situation with regard to the quality of sweat testing.6 The purpose of this editorial is to draw to the attention of paediatricans, and others requesting sweat tests, that they should be aware of the quality of the sweat test laboratory to which they refer their patients. See related paper by Mackay et al. on pp. 160–164. The sweat test was developed as a standardized clinical test in 1959 and has been the cornerstone of CF diagnosis ever since.1 Despite recent advances in genetic diagnosis and the recognition of an expanded CF phenotype, the sweat test has held up as a useful measure of activity of the affected protein, the cystic fibrosis transmembrane conductance regulator.7 In 98% of cases the sweat test gives a clear answer regarding the diagnosis, or exclusion, of cystic fibrosis. Interpretation of the results can be complicated when intermediate or borderline values are reported, but accurate interpretation of any result can only be made on the assumption that the test has been performed correctly. There are many ways to jeopardize the performance of a sweat test and the starting place to avoid this is by having a laboratory (and an individual at the laboratory) perform sweat tests regularly. National and international guidelines recommend laboratories perform 50 tests per annum and individuals 10 tests per annum. The details of every aspect of sweat testing can be found in the referenced guidelines, but key points include: sweat stimulation by pilocarpine iontophoresis, collection over a set area and time by either pad/gauze or Wescor Macroduct (Logan, Utah, USA), correct weight or volume of sweat collected and analysis of chloride as the principle analyte. Insufficient collections should not number more than 5% of the total sweat tests performed. Laboratories offering sweat tests should be actively engaged in a quality assurance program that is usually based on analysis of test solutions (RCPA-AACB currently offers this to Australian laboratories). These points form the basis of some simple questions that a clinician could ask to assess whether a laboratory was able to perform sweat tests reliably. The paper by Mackay et al. is disturbing because none of the 12 sweat test laboratories in New Zealand conformed to the guidelines developed in Australasia, the UK or the USA. Inconsistencies were observed at every point from sweat collection, analysis, reporting and interpretation of results. There did not appear to be anyone overseeing the performance of laboratories in New Zealand, a key point of the paper. Furthermore, there was no comment about the relationship between the national newborn screening program and CF care centres who are key stakeholders in the use of sweat test laboratories and may have a role in regulation of the laboratories. Involvement of the newborn screening program is essential, even though only 20% of sweat tests derived from screening, as this group of infants can be difficult to test. Screened infants are often asymptomatic so that the sweat test is critical to determining the diagnosis. Getting it wrong after newborn screening (both false positives and false negatives) is a big deal. CF care centres are clearly relying on the information given to them by the sweat test laboratories and must be sure the best quality sweat test has been performed. The questionnaire sent to all the sweat test laboratories in New Zealand will no doubt prompt some thinking as to the quality of their sweat tests, however, more is needed to correct this unfortunate state of affairs. Mackay et al. offer some solutions to the problems in New Zealand, but do they go far enough? Their solutions include closer adherence to the guidelines and the possibility of local collection of sweat with reference laboratories performing the analysis for regional areas without the facility for sweat analysis. The publishing of guidelines alone rarely has much impact, as the Mackay et al. paper highlights.8,9 There needs to be somebody with authority implementing the guidelines and controlling the outcomes. Perhaps the Australasian Association of Clinical Biochemists who have a set of published local sweat test guidelines need to take a lead on implementation and in New Zealand a regulatory authority needs to enforce these. I do not necessarily agree with the recommendation of local collection and reference laboratory analysis of the sweat sample. This only gets around the problems associated with sweat analysis but does not at all deal with the difficulties that can be encountered with sweat collection. Furthermore, there are little data to support the transport of sweat specimens. If delay in analysis is expected, the guidelines recommend maintaining the sample at 4°C in air-tight containers. Maintaining the ‘cold-chain’ could pose a problem. This particular recommendation by Mackay et al. suggests a compromise in ‘a small and scattered population who face significant geographical barriers to travel’.5 Things may not be much better in Australia with regard to the performance of sweat tests at inexperienced centres, although infants from regional centres identified by newborn screening who require a sweat test can be supported to travel to a centre with a reliable sweat test laboratory (state-based Isolated Patient Travel and Accommodation Schemes). Other patients who require a sweat test could also benefit from this scheme. It could be argued that for many patients the sweat test is one of the most important that they will have and they should be prepared to travel to a reliable laboratory. Clinicians in Australia and any country where CF is regularly considered should know what their sweat test laboratories are doing. Local guidelines, if not already developed, should follow the excellent UK and National Committee for Clinical Laboratory Standards ones and be implemented under some regulatory authority. If the sweat test is worth doing, it is worth doing properly. The author wishes to thank A/Prof Colin Robertson (Department of Respiratory Medicine, RCH) and Ms Ronda Greaves (Department of Complex Biochemistry, RCH) for reviewing the manuscript.