The impact of chest computed tomography and chest radiography on clinical management of cystic fibrosis lung disease.

Background: Recent standards of care mention chest radiography (CR) but not chest computed tomography (CT) in routine annual follow-up of children with cystic fibrosis (CF). To minimise radiation risk, CT or CR should only be performed if they impact clinical decision making. We investigated whether in addition to a wide range of commonly used clinical parameters, chest CT and/or CR in routine follow-up of CF patients influence clinical decisions.

Methods: 36 web based clinical vignettes (i.e. case simulations) were designed using clinical data from patients aged 8-18 years, randomly selected from two CF centres in The Netherlands. In a randomized cross-over design, clinicians assessed eight vignettes and suggested therapeutic/diagnostic management on two occasions, with a ten-week interval. Radiological information (CT or CR) was included at only one of the two assessments, in random order. Any differences in management could be attributed to information from CT or CR, and were compared by McNemar analysis.

Results: 44 European and Australian clinicians completed a total of 143 CT vignette pairs and 167 CR vignette pairs. CT was associated with a significant increase in antifungal treatment (Risk Ratio (RR) 2.8 (1.3-6.0, p = .02)), bronchoscopies (RR 1.6 (1.1-2.5, p = .04)), mycobacterial cultures (RR 1.3 (1.0-1.5, p = .02)), and 'need for hospitalization' (i.e. intravenous antibiotics and/or bronchoscopy) (RR 1.4 (1.0-1.9, p = .03)). CR led to a significant increase in inhaled antibiotics only (RR 1.3 (1.0-1.6, p = .04)).

Conclusions: CT but not CR, at routine biennial follow-up was associated with several changes in treatment and/or diagnostic testing, including the need for hospitalization.