Prospective evaluation of aspiration needle, cutting needle, transbronchial, and open lung biopsy in patients with pulmonary infiltrates.

Twenty consecutive patients with pulmonary infiltrates undiagnosed by routine, noninvasive methods were entered into a prospective study designed to evaluate the diagnostic yield of four methods of lung biopsy. Percutaneous aspiration needle, cutting needle, transbronchial, and open (anterior thoracotomy) biopsy were performed synchronously on all patients. Specimens were evaluated by microbiological, virological, and pathological methods. The diagnostic yields of the four methods were as follows: aspiration needle, 29%; cutting needle, 53%; transbronchial, 59%; and open lung biopsy, 94%. Open lung biopsy was significantly better in yielding a diagnosis than aspiration needle (p less than 0.001), cutting needle (p less than 0.001), and transbronchial biopsy (p less than 0.04).