CT FINDINGS IN BRONCHIECTASIS - LIMITED VALUE IN DISTINGUISHING BETWEEN IDIOPATHIC AND SPECIFIC TYPES

OBJECTIVE. The purpose of this study was to determine whether the patt ern and distribution of bronchiectasis shown on CT scans can be used t o discriminate between idiopathic cases and those with an identifiable cause. MATERIALS AND METHODS, The CT scans of 168 patients with chron ic purulent sputum production and who were suspected of having bronchi ectasis were analyzed (117 patients with idiopathic bronchiectasis, 15 with allergic bronchopulmonary aspergillosis, 15 with hypogammaglobul inemia, 15 with impaired mucociliary clearance, and seven with cystic fibrosis diagnosed in adult life). The scans were analyzed in random o rder by two observers. The extent, site, type, and lobar distribution of bronchiectasis and the severity of bronchial dilatation and bronchi al wall thickening were scored, The frequency of these features in the known-cause groups was compared with that in the idiopathic group to identify any significant differences. RESULTS. Compared with idiopathi c bronchiectasis, no significant lobar predominance was seen in any of the known-cause groups, apart from a higher frequency of lower lobe i nvolvement in the patients with syndromes of impaired mucociliary clea rance (p < .02). The bronchiectasis of allergic bronchopulmonary asper gillosis and adult cystic fibrosis was more often widespread (five or six lobes involved (p < .001 and p < .01, respectively) than idiopathi c bronchiectasis, Central bronchiectasis was more common in allergic b ronchopulmonary aspergillosis (p < .005), although the sensitivity whe n this was used as a diagnostic feature was only 37%, In all groups, c ylindrical bronchiectasis was the most common type, with varicose and cystic bronchiectasis occurring more frequently in allergic bronchopul monary aspergillosis (p < .01). On multiple regression analysis, aller gic bronchopulmonary aspergillosis and adult cystic fibrosis showed mo re extensive disease than idiopathic bronchiectasis (p < .0005 and p < .001, respectively), independent of other CT features. In hypogammagl obulinemia, dilatation of the bronchial lumen was less than in idiopat hic bronchiectasis (p < .02) independent of disease extent and bronchi al wall thickness. CONCLUSION, Although differences in distribution an d morphology of bronchiectasis may be seen on CT scans in groups of pa tients with bronchiectasis of different causes, CT findings applied to individual patients are of limited value in discriminating between id iopathic bronchiectasis and bronchiectasis of various known causes.