Introduction Antifibrotic drugs for idiopathic pulmonary fibrosis (IPF) patients in England and Scotland are only available to those with FVC percent predicted (FVC%pred) less than or equal to 80%. The prescribing guidance does not state which reference values should be used.

Aims

1. To find out if the use of different sets of reference values affects the numbers of patients with FVC%pred greater than and less than 80%.

2. To find out which reference equations were in use at UK centres prescribing antifibrotics for IPF in April 2016.

Methods We searched databases for patients diagnosed with IPF at our interstitial lung disease (ILD) unit from 1/1/2010 to 31/12/2015. We calculated FVC%pred using three different sets of reference values (ECSC, GLI or NHANES). The chief respiratory physiologist in each ILD centre in England was contacted and asked which reference values they used to calculate FVC%pred. In Scotland, four hospitals with an ILD MDT were contacted and asked the same. McNemar tests were used to compare the proportion of patients eligible for antifibrotic prescription when FVC%pred was calculated by ECSC or either NHANES or GLI.

Results See Table 1. We identified 671 unique patients: after exclusions, 528 had complete data.

There was a higher proportion of patients calculated to have an FVC%pred >80% (ineligible for antifibrotics) using ECSC than GLI: Chi-square 22.0, 1df, P<0.0001. The difference in proportions was greater when ECSC was compared to NHANES: Chi square 33.03, 1df, P < 0.0001. Of 30 patients with ECSC FVC%pred 80–85%, 27 [90%: 95% CI: 79–100%) had their FVC%pred fall to <80% when recalculated with NHANES.

18 of 20 ILD centres in England were using ECSC to calculate FVC%pred; others used GLI (n = 1) and Falaschetti (n = 1). All four Scottish centres were using ECSC.

Discussion Many patients with ECSC FVC%pred too high for antifibrotics fall into the eligible range when NHANES, the set of reference values used in the ASCEND pirfenidone trial, or GLI, as recommended by the UK Association for Respiratory Technology and Physiology(ARTP) are used.

Conclusions We urge physicians and physiologists to ensure that reference values used to calculate FVC%pred are cited in lung function reports. Those choosing reference values must be aware of implications for patients.