Purpose: To review autopsy-proven cases of opportunistic pneumonia and determine how many of these patients had received corticosteroid therapy for obstructive lung disease in order to define whether this therapy was the major risk factor predisposing to infection.
Patients and methods: All autopsies performed at Winthrop-University Hospital over a 5-year period were reviewed, and 30 cases of opportunistic pneumonia were identified. In eight of 30 cases, corticosteroid therapy for chronic obstructive pulmonary disease (COPD) was the only identifiable risk factor for opportunistic infection. The other 22 patients had other well-defined risk factors for infection. Chart review of the eight patients with COPD was undertaken to define the clinical features of their infections.
Results: All eight patients had a progressive multilobar pneumonia that failed to resolve, either clinically or radiographically, despite the use of multiple broad-spectrum antibiotics. In four cases, the infection was community-acquired, while in the other four cases, it was nosocomial in origin. Despite the presence of a nonresolving pneumonia, opportunistic infection was generally not considered as a diagnostic possibility, with only one case being correctly diagnosed antemortem. Autopsy examination documented Aspergillus species as being responsible for six episodes of pneumonia, Candida albicans accounting for one episode, and cytomegalovirus accounting for one episode.
Conclusion: Based on this experience, it is clear that corticosteroid therapy of COPD can lead to opportunistic pulmonary infection, in or out of the hospital. This diagnosis should be considered when patients receiving this therapy develop a pneumonia that fails to respond to broad-spectrum antibiotics.