Idiopathic Pulmonary Fibrosis Prognosis Predicted by CT Parameters

Patients with idiopathic pulmonary fibrosis (IPF) with better computed tomography (CT) indices such as forced expiratory pressure in 1 second (FEV₁) and mean lung attenuation had improved outcomes, according to the results of study published in CHEST.

Lung function parameters of patients recently diagnosed with IPF were assessed after segmentation of the lungs in thin-section CT in the supine position at full inspiration. Parameters of interest measured were mean lung attenuation, skewness, kurtosis, peak attenuation, total lung area, inflexion point with slope, and the area right of the inflexion point (AROIP). Correlations were then made between prognosis and lung function at diagnosis.

Of the 70 study participants (mean age, 66.4 years), 48 were men and22 were women. The strongest correlation was between mean lung attenuation and FEV₁. Diffusing lung capacity had the least correlation with histogram features, except for the AROIP.

Importantly, researchers analyzed the predictive performance of the 4best histogram features (mean lung attenuation, skewness, kurtosis and AROIP) for each of the 5 outcomes (physiologic measurements). These analyses indicated that mean lung attenuation and AROIP were the best single predictors of physiologic outcome. With respect to survival, 55.7% of the participants died during the follow-up period. AROIP was the most significant single predictor of mortality (P ≤.001).

“The use of the newly defined AROIP might be valuable for prediction of [diffusing capacity for carbon monoxide] and mortality,” the researchers wrote. “We therefore think that CT indices obtained at the time of diagnosis can provide valuable prognostic information.”

Disclosures: Several authors report financial relationships with pharmaceutical companies.

Reference

Loeh B, Brylski L, von der Beck D, et al. Lung CT densitometry in idiopathic pulmonary fibrosis (IPF) for the prediction of natural course, severity and mortality [published online February 8, 2019]. CHEST. doi:10.1016/j.chest.2019.01.019