Idiopathic Interstitial Pneumonias: Diagnostic Accuracy of Thin-Section CT in 129 Patients

Takeshi Johkoh(Vancouver Hospital and Health Sciences Centre), Néstor L. Müller(Vancouver Hospital and Health Sciences Centre), Yannick Cartier(Vancouver Hospital and Health Sciences Centre), Peter V. Kavanagh(Vancouver Hospital and Health Sciences Centre), Thomas E. Hartman(Mayo Clinic in Arizona), Masanori Akira(National Kinki Chuo Hospital for Chest Disease), Kazuya Ichikado(Kumamoto University), Masayuki Ando(Kumamoto University), Hironobu Nakamura(Osaka University)
Cited by 362

Abstract

PURPOSE: To determine whether idiopathic interstitial pneumonias can be differentiated on the basis of the pattern and distribution of abnormalities at thin-section computed tomography (CT). MATERIALS AND METHODS: Thin-section CT scans in 129 patients with histologically proved idiopathic interstitial pneumonia (35 with usual interstitial pneumonia [UIP], 24 with bronchiolitis obliterans organizing pneumonia [BOOP], 23 with desquamative interstitial pneumonia [DIP], 20 with acute interstitial pneumonia [AIP], and 27 with nonspecific interstitial pneumonia and fibrosis [NIPF]) were independently assessed by two observers without knowledge of clinical or histologic data. The observers recorded the abnormalities, diagnosis, and degree of confidence in their diagnosis. Differential diagnosis was limited to the five types of idiopathic interstitial pneumonia. RESULTS: The two observers made a correct diagnosis, on average, in 74 (57%) cases. On average, the correct diagnosis was made in 25 (71%) cases of UIP, 19 (79%) of BOOP, 14.5 (63%) of DIP, 13 (65%) of AIP, and 2.5 (9%) of NIPF. The two observers made a correct diagnosis with a high degree of confidence in 50 (39%) readings. There was moderate agreement between the observers for the correct diagnosis (k = 0.55) and for the correct diagnosis with a high degree of confidence (k = 0.65). CONCLUSION: Except for NIPF, the various subtypes of idiopathic interstitial pneumonias often have a characteristic appearance that allows differentiation at thin-section CT.


Related Papers

No related papers found

Powered by citation graph analysis