Computed tomography of the pulmonary parenchyma. Part 2: Interstitial disease

Elias A. Zerhouni, David P. Naidich, Frederick P. Stitik, Nagi F. Khouri, Stanley S. Siegelman

Research output: Contribution to journalArticlepeer-review

121 Scopus citations


A series of patients with documented predominantly interstitial pulmonary disease was examined by routine and high-resolution computed tomography (CT) and compared to a series of twenty-one normals. Inspiratory-expiratory lung density measurements were also obtained at predetermined levels. Several basic CT signs of interstitial disease were identified: (1) finely irregular and thickened pleural surfaces; (2) irregular vascular shadows; (3) thickened and irregular bronchial walls making bronchi visible over a longer portion of their course in the lungs; (4) reticular network of lines with three patterns easily distinguishable by the size of their reticular element; (5) hazy patches of increased density of various sizes distinguishable from alveolar filling processes by the fact that vessels can still be visualized through them; and (6) nodules of various sizes. Micronodules are often associated with a small or medium-size reticular network and in most cases seem to represent points of confluence rather than isolated nodules. The hematogenous origin of some nodules can be specifically suggested when feeding vessels arc demonstrated on thin-section scans. Nodules associated with a large network of thickened septa are suggestive of lymphangitic carcinomatosis. Inspiratory-expiratory density gradients can be more useful in confirming the diagnosis of interstitial disease than absolute measurements.

Original languageEnglish (US)
Pages (from-to)54-64
Number of pages11
JournalJournal of thoracic imaging
Issue number1
StatePublished - Dec 1985

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Pulmonary and Respiratory Medicine


Dive into the research topics of 'Computed tomography of the pulmonary parenchyma. Part 2: Interstitial disease'. Together they form a unique fingerprint.

Cite this