Our patient recovered from the acute episode of CMV pneumonia with the combined therapy of GIG and DHPG. Although CMV was not detected in the BAL fluid after the therapy, lymphocytosis was found in the BAL during the acute infection and persisted after recovery in the absence of detectable viral particles. This provides strong evidence that CMV infection results in the activation of the host’s immune system, as proposed by Grundy et al. Furthermore, it points out that the two-week combined therapy does not result in the cessation of the activated host response, which is the proposed pathologic process. The clinical significance of persistent lymphocytosis is not clear at the moment; however, it probably played a major role in the development of BOOP in this patient. The pathogenesis of BOOP is not known; however, similar to other immunopathologic conditions, most cases with BOOP are responsive to steroid therapy. Epler et al reported complete recovery in 65 percent of their patients treated with corticosteroids. Furthermore, Chan et al suggested that bronchiolitis obliterans in patients with BMTs may be due to an activated immune response from GVHD or a joined interaction of a viral infection and GVHD. Thus, it seems likely that the BOOP in our patient was associated with CMV pneumonia and related to the persistent lymphocytosis in the BAL fluid.
This case report suggests that CMV pneumonia in patients with allogeneic BMTs should be divided into two phases, an acute disease and a chronic sequela. In the acute phase, fulminant interstitial pneumonitis carries a high mortality. The combined therapy of GIG and DHPG may control the disease in a fair proportion of cases at this stage; however, the two-week therapy may not result in the cessation of the immunopathologic process, which may lead to serious im-mune-mediated pulmonary complications such as the BOOP in our patient. Thus, the role of maintenance therapy in the prevention of chronic complications needs to be addressed in future studies.