BACKGROUND: Cryptogenic fibrosing alveolitis (CFA) is an uncommon disorder of unknown aetiology characterised by interstitial fibrosis which typically shows a restrictive pattern on pulmonary function testing. Some patients with CFA and relative preservation of lung volumes have been described and it has been suggested that their volume preservation may be due to concomitant emphysema. In a retrospective study the relative frequency of preserved lung volumes in CFA, its relationship to emphysema determined by CT scanning, its clinical features, and its subsequent natural history were investigated. METHODS: Using predefined characteristics 48 patients with CFA were identified from pulmonary function records over a three year period. Volume preservation was defined as a forced vital capacity (FVC) of > 80% predicted at presentation. Patients with relative volume preservation were compared with those with more typical pulmonary restriction and clinical data at presentation, and details of their subsequent prognosis, treatment and loss of lung function with time were obtained. Where available, computed tomographic (CT) scans for the two groups were compared in a blinded fashion to score the extent of fibrosis and the presence of concomitant emphysema. RESULTS: Twenty one (44%) of the patients with CFA had a FVC of > 80% predicted. They were more likely to be male (76% versus 48%) and to be current smokers (57% versus 22%) with a heavier life time cigarette consumption than the restricted patients (mean (SE) 38 (4.6) versus 25 (4.5) pack years). There were no significant differences in prognosis and subsequent treatment between the groups. Comparable HRCT scans were available in 23 subjects (seven preserved, 16 restricted). They showed no difference in extent of the pulmonary fibrosis but patients with volume preservation were more likely to show concomitant emphysema (86% versus 19%). Patients with emphysema on HRCT scans were heavier smokers (41(10) versus 21(17) pack years) than those without emphysema but there was no difference in the extent of CFA score between the two groups. CONCLUSIONS: In this area of high smoking prevalence a significant number of patients with CFA presented with relative preservation of lung volumes and FEV1/FVC ratio. In many of these subjects this appears to reflect coincidental emphysema. This may make interpretation of gas transfer factor used to monitor progression in CFA difficult. However, there was no evidence that lung volumes at presentation were of prognostic significance.