Lung-Dominant Connective Tissue Disease: Clinical, Radiologic, and Histologic Features.

BACKGROUND: Lung-dominant connective tissue disease (LD-CTD) is a disease concept for interstitial pneumonia; however, it has not been robustly validated. This study was conducted to elucidate the clinical, radiologic, and histologic features of LD-CTD.

METHODS: We retrospectively reviewed 44 consecutive patients with serologically defined LD-CTD who underwent surgical lung biopsy. Patients were identified as having LD-CTD if they had specific autoantibodies but did not meet the criteria for connective tissue disease. We conducted a multidisciplinary diagnosis and evaluated major histologic patterns according to the current idiopathic interstitial pneumonias (IIPs) classification of 2013. Characteristic histologic features for LD-CTD (eg, prominent plasmacytic infiltration, lymphoid aggregates with germinal centers), high-resolution CT (HRCT) scan patterns, and prognosis were also assessed.

RESULTS: The major histologic patterns were usual interstitial pneumonia (UIP) in 25 patients and nonspecific interstitial pneumonia (NSIP) in 13 patients. Two or more characteristic histologic features for LD-CTD were observed in 15 patients with histologic UIP (h-UIP) and 11 patients with histologic NSIP (h-NSIP). Fifteen patients with h-UIP (60%) showed an inconsistent UIP pattern on HRCT scan. After multidisciplinary discussion (MDD), 18 patients with h-UIP were labeled as having unclassifiable IIP. The annual change in percent predicted FVC improved significantly in patients with h-NSIP (P = .002), who also had better survival than those with h-UIP (P = .031). In contrast, survival was not associated with HRCT scan pattern (P = .79).

CONCLUSIONS: The major histologic patterns in LD-CTD were UIP followed by NSIP. Two-thirds of patients had characteristic histologic features for LD-CTD. A majority of patients with h-UIP were considered to have unclassifiable IIP based on MDD. Patients with h-UIP had worse survival than those with h-NSIP.