Dear Sirs,

We were very interested to read the correspondence from Gruffydd-Jones et al. in the most recent issue of the PCRJ.1 We thank them for their request that we should provide further clarification regarding within-session repeatability when performing diagnostic spirometry.

The 2009 Spirometry Standards document to which they refer2 recommended 150 ml as the limit for within-session repeatability for FEV1 and FVC, in accordance with American Thoracic Society (ATS) and European Respiratory Society (ERS) standards. Gruffydd-Jones et al. favour a target of 100 ml.1 However, apart from reference to previous correspondence from Fletcher and Loveridge3 (citing a sample of 10 subjects), and Cooper,4 they provide no published evidence to support their opinion. The ATS/ERS guidelines5 have not been updated (the 150 ml limit remains current), nor has there been an update to the British Thoracic Society (BTS)/Association for Respiratory Technology and Physiology (ARTP) guideline published in 1994 which suggested 100 ml.6

Nevertheless, we note that Gruffydd-Jones et al. point out that the GOLD guideline recommendation in 20117 moved to 100 ml. We have followed-up this point. In fact, this is apparently a typing error (personal communication from Jorgen Vestbo, Chair GOLD Science Committee); on page 12, the guidance suggests “5% or 100 mL whichever is the greater”. In effect, if the patient has an FEV1 or FVC of more than 2 litres the 5% guidance takes precedence and makes the 100 ml reading redundant.

In primary care we aim to achieve comparable standards to our specialist colleagues. Ferguson et al., in their consensus statement from the National Lung Health Education Program (NLHEP), suggested a rating system (A-F) for assessing quality of spirometry.8 Grades A and B required a minimum difference of 100 and 101–150 ml respectively, with Grade C requiring 151–200 ml difference between the best two FEV1 and FVC readings. Grade D required only one acceptable manoeuvre but with FEV1 values within 200 ml, and Grade F signified no acceptable manoeuvres.

Three recent publications have utilised a similar quality control grading system, involving over 55,000 spirometry tests by specialists and in primary care, and demonstrate quite clearly that the 100 ml limit suggested by Gruffydd-Jones and colleagues is unrealistic.

Enright et al.9 studied 13,599 good quality spirometry tests by specialists at the World Trade Center, 80% of which achieved grade A and B standards (within 200 ml). Leuppi et al. considered 29,817 consecutive spirometry tests that had taken place in primary care and found that 41% achieved grade A and B (within 200 ml) and 11.8% had the lowest grade F.10 Finally, recently published data from the European Spirometry Tent performed at ERS meetings,11 which reported on 12,448 tests of which 10,395 (83.5%) were termed acceptable (only grade F rejected), showed that the overall standard for grade A and B in this specialist environment was 30.8% (with the best results being undertaken in 2004 in Glasgow, but still only achieving 51.4%).

As the available evidence suggests that our specialist colleagues find it difficult to achieve repeatability within 200 ml in 50% of tests, it is inappropriate to suggest setting a 100 ml standard for within-session repeatability in general practice. We therefore maintain our recommendation — i.e. that spirometry within-session repeatability should be within 150 ml in keeping with ATS/ERS guidance, until justified by evidence from specialists to the contrary.

However, we strongly support efforts like the European Spirometry Driving License project aimed at improving standards of measurements, which will in the future enable the adoption of stricter quality criteria in general as well as in specialist practice, and we look forward to the publication of high quality evidence demonstrating that this level of measurement is achievable.