RT Journal Article
SR Electronic
T1 Connective tissue disease related fibrotic lung disease: high resolution computed tomographic and pulmonary function indices as prognostic determinants
JF Thorax
JO Thorax
FD BMJ Publishing Group Ltd and British Thoracic Society
SP 216
OP 222
DO 10.1136/thoraxjnl-2013-203843
VO 69
IS 3
A1 Simon L F Walsh
A1 Nicola Sverzellati
A1 Anand Devaraj
A1 Gregory J Keir
A1 Athol U Wells
A1 David M Hansell
YR 2014
UL http://thorax.bmj.com/content/69/3/216.abstract
AB Purpose To determine high resolution computed tomography (HRCT) patterns and pulmonary function indices which are associated with increased mortality in patients with connective tissue disease related fibrotic lung disease (CTD-FLD). Methods HRCTs from 168 patients with CTD-FLD were scored by 2 observers for a variety of HRCT patterns and traction bronchiectasis. A radiological diagnosis of usual interstitial pneumonia (UIP), fibrotic non-specific interstitial pneumonia (NSIP) or indeterminate was also assigned. Using Cox regression analysis, associations with mortality were identified. Honeycombing and traction bronchiectasis scores were converted to binary absence/presence scores and also tested. A subgroup analysis of patients with biopsy material (n=51) was performed by classifying patients according to radiological and histopathological diagnoses, as concordant UIP, discordant UIP and fibrotic NSIP. The prognostic separation of this classification was also evaluated. Results Severity of traction bronchiectasis (HR 1.10, p=0.001, 95% CIs 1.04 to 1.17), increasing extent of honeycombing (HR 1.08, p=0.021, 95% CI 1.03 to 1.13) and reduction in DLco (HR 0.97, p=0.013, 95% CI 0.95 to 0.99) were independently associated with increased mortality. Interobserver agreement and prognostic strength were higher for binary traction bronchiectasis scores (weighted κ (κw)=0.69, HR 4.00, p=0.001, 95%CI 1.19 to 13.38), than binary honeycombing scores (κw=0.50, HR 2.87, p=0.022, 95% CI 1.53 to 5.43). The radiological-histopathological classification was strongly associated with increased mortality (HR 2.74, p<0.001, 95% CI 1.57 to 4.77) and patients with discordant UIP had a better prognosis than concordant UIP but worse prognosis than fibrotic NSIP. Conclusions Severity of traction bronchiectasis, extent of honeycombing and DLco are strongly associated with mortality in CTD-FLD. Interobserver agreement for traction bronchiectasis is higher than for honeycombing. In CTD-FLD, radiological diagnosis has survival implications in biopsy proven UIP.