Prediction of IPF within 1–2 year with the early changes in quantitative imaging patterns using high resolution computed tomography

Abstract

IntroductionHigh resolution computed tomography (HRCT) plays an indispensable role in the diagnosis of Idiopathic Pulmonary Fibrosis (IPF). Due to the short median survival of 2 to 5 years, understanding the chance of progression in the next follow-up of a patient is critical. The aim of this study is to test the predictability of progression in IPF of the near follow-up of a patient using the early quantitative changes in HRCT scans. MethodsVolumetric anonymized HRCT scans were obtained. Two data sets were collected: a pilot study with 35 subjects in three sequential scans (baseline, 6 and 12 month scans) and validation study involving 157 subjects with clinical information. Quantitative lung fibrosis (QLF) scores were calculated for HRCT scans using an automated system. A log-rank test and Cox regression were used to test a prediction with early changes in QLF. ResultsIn the pilot study mean changes in QLF increased by 4.33% at 6 months and 14.83% at 12 months, when the visual assessments changed from stable to worse. Mean QLF changes were ~1% for the stable cases during 12 months. Using 4% QLF differences at 6 months in the validation cohort, significant differences in progression were found (p < 0.001). ConclusionEarly changes in a quantitative score from HRCT can predict progression in lung function. A quantitative texture-based score can play a pivotal role for informed and timely management decision for patients with IPF.