From “traction bronchiectasis” to honeycombing in idiopathic pulmonary fibrosis: A spectrum of bronchiolar remodeling also in radiology?

By CT, a “definite” usual interstitial pneumonia (UIP) pattern, as seen in IPF, is characterized by the presence of reticulation, traction bronchiectasis and honeycombing in a basal and peripheral predominant distribution. The presence of honeycombing and traction bronchiectasis, besides reticulation, is crucial [12‐19].

However if we reconsider morphological aspects in light of the pathogenic events discussed above, we identify some interesting key points in the interpretation of CT findings.

First, the sites in which mechanical stress and remodeling are highest, correspond to the areas in which traction bronchiectasis and honeycombing appear.

Second, most of the “scarred” tissue is in the region distal to the traction bronchiectasis, beneath the pleura, and does not concentrically surround the dilated bronchi. Conversely in nonspecific Interstitial Pneumonia (NSIP), traction bronchiectasis is completely surrounded by the fibrotic tissue (Fig. 1).

Furthermore as the fibrosis progresses, dilatation of the airways increases in severity from the periphery to the inner third of the lungs, passing from a score 1 (more peripheral) to 3 (from periphery to the inner third of the parenchyma) [score proposed by Edey et al. [20]] (Fig.2).

In CT scans, when fibrosis is more severe, bronchiectasis tend to follow the convexity of the pleura and to overlap with honeycombing features [21].

Therefore, also in CT scans, traction bronchiectasis observed in IPF subjects is better interpreted as resulting from bronchiolar proliferation rather than from pure mechanical traction of a single airway by scarring tissue. Supporting this point of view is the recent observation by Staats et al. [4] on explanted lungs obtained from patients with IPF. They showed a positive correlation between honeycombing assessed by CT and bronchiolectasis (p = 0.001) and respiratory-lined cysts (p = 0.001) histologically counted. If we consider traction bronchiectasis as the solely consequence of pure mechanical traction around the airway, this process should result in a relatively stable number of traction “holes” with an enlargement of lumen reaching the periphery, as actually is present only in NSIP. Thus, in IPF, remodeling process appears to be a continuum from traction bronchiectasis to honeycombing and conceptual separation of the two processes may be misleading.

Walsh et Al. [19] reinforced this concept recently. The authors retrospectively reviewed radiological features of 162 biopsy proven cases of UIP and NSIP, delineating a radiological visual score for each case and correlating these radiological data to the fibroblastic foci profusion score and other morphologic aspects. They concluded that in UIP there was a strong correlation between traction bronchiectasis, honeycomb changes and fibroblastic foci profusion. In conclusion, according to the “alveolar stem cell exhaustion” model explaining at least part of the pathogenic events in IPF [14] we suggest that traction bronchiectasis and honeycombing is a unique and continuous process of bronchiolar dysplastic proliferation and to interpret accordingly the HRCT features.