Pathology of Granulomatous Pulmonary Diseases

Abstract

Context.— Because granulomas are represented in almost every disease category, the number of clinically and pathologically important granulomatous pulmonary diseases is large. Their diagnosis by pathologists is particularly challenging because of their nonspecificity. A specific diagnosis can be achieved only when a granuloma-inciting agent(s) (eg, acid-fast bacilli, fungi, foreign bodies, etc) are identified microscopically or by culture; this does not occur in most cases. Furthermore, a specific diagnosis cannot be reached in a high percentage of cases. Although sarcoidosis and infectious diseases account for approximately half of pulmonary granulomatous diseases worldwide, there is significant geographic variation in their prevalence. Objectives.— To present updated information to serve as a guide to pathologic diagnosis of pulmonary granulomatous diseases, to address some commonly held misconceptions and to stress the importance of multidisciplinary coordination. Presentation of basic aspects of granulomas is followed by discussion of specific disease entities, such as tuberculous and nontuberculous Mycobacterial infections, fungal, bacterial, and parasitic infections, sarcoidosis, necrotizing sarcoid granulomatosis, berylliosis, hypersensitivity pneumonitis, hot tub lung, rheumatoid nodule, bronchocentric granulomatosis, aspirated, inhaled, and embolized foreign bodies, drug-induced granulomas, chronic granulomatous disease, common variable immunodeficiency, and granulomatous lesions associated with various types of cancer. Data Sources.— Review of pertinent medical literature using the PubMed search engine and the author's practical experience. Conclusions.— Although the diagnosis of granulomatous lung diseases continues to present significant challenges to pathologists, the information presented in this review can be helpful in overcoming them. The importance of multidisciplinary coordination in cases where morphologic diagnosis is not possible cannot be overstated.