Objective: To determine the diagnostic yield achieved with the application of current recommendations for evaluating patients with suspected interstitial lung disease (ILD) and the procedures that must be applied to reach a definitive diagnosis.
Patients and methods: Over a 10-year period, 500 consecutive patients attending an ILD outpatient clinic who showed features of diffuse lung involvement were assessed with a single diagnostic protocol. Results were introduced in a dedicated database and diagnoses for idiopathic interstitial pneumonia were established according to a recent consensus classification.
Results: A definitive diagnosis was reached in 427 (85%) patients: in 125 without invasive procedures and in 302 with invasive procedures. In 73 (14.6%) cases a definitive diagnosis was not reached, and patients were placed in the group of unclassifiable interstitial pneumonia. Idiopathic interstitial pneumonia was the predominant group with 193 (39%) patients. The main specific entities included sarcoidosis with 93 (19%) patients, usual interstitial pneumonia with 84 (17%) patients, and hypersensitivity pneumonitis with 75 (15%) patients. Thirty (6%) patients were diagnosed with an illness other than ILD (false ILD). In 332 patients, we performed a total of 433 invasive procedures: transbronchial biopsy in 252 (direct diagnostic yield, 38%, or if used also to exclude other specific diagnosis, 50%), bronchoalveolar lavage in 260 (yield, 5%), and open lung biopsy in 141 (yield, 93%). Hence, following the current diagnostic approach, a definitive diagnosis was established for 85% of patients, for 25% solely on clinical grounds and imaging criteria and for 60% on the basis of invasive procedures. Diagnosis by open lung biopsy was still required for 141 (28%) patients.
Conclusions: The diagnostic yield was high when the recommended study protocol was followed. A quarter of the diagnoses were reached with clinical criteria alone, but another quarter could only be made after open lung biopsy.