Objective clinical scoring system and diagnostic variability in idiopathic pulmonary fibrosis

The diagnosis of idiopathic pulmonary fibrosis (IPF) is predicated on high-resolution computed tomography (HRCT) patterns, elimination of a secondary cause of interstitial lung (ILD) and histopathology when necessary. The Fairfax IPF Clinical Score (FICS) integrates 8 parameters weighted to estimate the clinical probability of IPF within an ILD population. The aim of this study was to evaluate risk stratification for IPF by a diverse panel of independent clinicians in comparison to the FICS.

A series of 16 cases were randomly selected from a French ILD tertiary hospital (8 IPF and 8 non-IPF ILD), with their diagnoses established through multidisciplinary discussion. Demographic and clinical data were provided to 15 French and American pulmonologists, 8 of whom work in ILD tertiary care centers. They were asked to evaluate the probability of IPF for each case (defined as low < 30%, intermediate low 30-50%, intermediate high 50-70%, high > 70%) with clinical information only, then after integrating their reading of the HRCT. Their responses were analyzed, compared between subgroups and with the FICS.

For the 16 cases, IPF patients were numerically older, more frequently males, smokers, and Caucasian compared to non-IPF cases. Analysis of the demographic and clinical parameters led to a correct categorization of IPF patients as intermediate high or high likelihood in 73% of cases, with ILD experts appearing to perform better than non-ILD experts in this regard. The FICS was higher in IPF vs. non-IPF patients. The FICS scored IPF patients higher and non-IPF lower than clinicians respectively in 51% and 48% of cases. The contribution of imaging changed the assessment of IPF probability in only 16% of cases.

In this study, we have evaluated how integrating clinical data, alone or in an objective fashion, may be useful for IPF diagnostic accuracy. Our study emphasis that clinical data is important for an IPF diagnosis and the FICS appears to be an effective and reproducible tool that might enable diagnostic standardization between general pulmonologists and ILD experts. ILD experts seem more adept at diagnosing IPF, possibly due to their greater exposure and participation in MDDs. These results need further validation in larger prospective cohorts.