P. jirovecii is an extracellular fungus that almost exclusively inhabits alveolar spaces[15], causing pulmonary infection through adhesion to the surface of alveolar epithelium cells (type I pneumocytes) [16]. CMV belongs to the herpes viridae family, with seroprevalence of 40%–100% approximately in the world’s population [4, 17], which could establish life-long infection and reactivate during episodes of immunosuppression [18]. The cell-mediated immunodeficiency is particularly essential for the development of PCP and CMV-P. PCP (approximately 90% cases) usually occurs with CD4 lymphocyte counts < 200 cells/mm3, while CMV-P tends to occur with CD4+ T lymphocyte counts༜50 cells/mm3 [19]. The results of present study showed that CD4+ T lymphocyte counts of PCP group and CMV-P group were both lower than 50 cells/mm3 (22 and 22.50 cells/mm3,respectively), which were consistent with previous studies. Nevertheless, our data also demonstrated that the neutrophil percentage in PCP was higher than in CMV-P, while the lymphocyte percentage in CMV-P was higher than in PCP, this were consistent with previous studies as well [20].
The identification between PCP and CMV-P is particularly challenging since clinical characteristics are similar in both groups. As P. jirovecii is difficult to cultivate in vitro, the diagnosis usually established with identification of the pathogen in respiratory specimens (BAL fluid ideally or induced sputum otherwise) by sensitive immunocytochemical methods or polymerase chain reaction (PCR) [8–10]. In this study, we diagnosed PCP when clinical features were present combined with P. jirovecii were detected by silver stain or PAS in a BAL fluid. With regard to CMV-P, a diagnosis requires clinical symptoms of pulmonary infection combined with identification of CMV in lung biopsy or BAL fluid by rapid culture, viral isolation, immunohistochemistry, histopathology, or DNA hybridization techniques [11]. In the current study, the diagnosis of CMV-P were established when clinical evidence for infection coinciding with definitive BAL results of fluorescence quantitative PCR. However, Microbiological confirmation is usually invasive or difficult to obtain, which may have delayed the diagnosis and initiation of therapies. To identify the differences between these two pathogens in radiological features, we retrospectively compared the HRCT features in 78 patients with PCP and in 34 patients with CMV-P.
Our data showed that the frequency of GGO were significantly common, while the occurrence of cavity was relatively rare in both cohorts. Meanwhile, among patients with PCP and CMV-P, the frequency of reticulation, bronchial wall thickening, mosaic perfusion, cyst, crazy-paving pattern, TIB, and LN enlargement are similar to each other. These findings are in accordance with previous studies concerning infections in immunocompromised patients[21]. Kunihiro et al demonstrated that representative CT features in immunosuppressed patients included diffuse GGO with mosaic pattern in PCP and nodules with or without the halo sign in CMV-P[22]. PCP was reported to have variable atypical CT features; 18% of PCP patients accompany LN enlargement, 3%-5% show nodular lesions [23]. Concerning CMV-P, a mixed pattern with GGO, consolidation, and nodules was observed to its common CT features in previous studies of AIDS patients [21].This would suggest that there are several similarities in CT features between these two types of pneumonia.
In the current study, the comparisons made between PCP and CMV-P showed that the occurrence of consolidation, nodules, and the halo sign is more common in CMV-P than in PCP. In addition, large nodules with perilymphatic distribution were not found in any of patients with CMV-P group. In addition, the nodules’ occurrence were positively correlated with CD4+T cell count, CD8+T cell count and CD4/CD8 ratio in CMV-P patient, while no significant correlations were found in PCP patients. These similarities and differences in CT features may be correlated to the histopathology in each type of pneumonia. The primary histopathological feature of PCP and CMV-P is diffuse alveolar damage [24, 25]. GGO and reticulation represent the pathological features of diffuse alveolar damage [26, 27]. Thus, CT features of GGO and reticulation appearance are common in both groups. However, some studies suggest that the pulmonary injury of PCP is induced by exuberant inflammation[28], while the lung damage of CMV-P appears to be the direct result by the cytopathogenic effects of CMV in AIDS patients[29]. Consolidation refers to advanced stages of the inflammation in which the alveoli are filled with exudates or other product, rendering the lung solid[30]. Considering the differences in immune status and viral loads between the two cohorts, consolidation in patients with CMV-P should be more frequently observed than with PCP. Nodule was reported as a common CT finding in patients with CMV-P[29]. Pathologically, nodules corresponding to areas of hemorrhagic or inflammatory nodules indicate the intraalveolar aggregations of red blood cells, macrophages, and fibrin [29]. Hypotheses have been assumed to interpret the mechanism of rare nodules in PCP including microorganism factors, the absence of IgA Pneumocystis antibodies and change in CD4 T cell [31–33]. It is reported that the halo-sign was observed in approximately 60% of viral etiology cases. Finding multiple nodules smaller than 10 mm in appropriate clinical scenario have a sensitivity, specificity, and negative predicted value for viral infection of greater than 80%[34].
There are several limitations in present study. The patients included were in different stages of diseases because of using the BAL results as diagnostic criteria, which might influenced the observed frequency of CT features, such as halo signs and cavities. The lack of histological specimens limited the analysis of correlation with pathologic radiology. The analysis of correlation between histology and pathologic radiology was limited owing to the lack of histological specimens. In addition, sensitivity and specificity of these patterns in AIDS patients should be further evaluated in a blinded retrospective or future prospective study.