PURPOSE: To analyze the capabilities of chest roentgenogram (CXR) and computed tomography (CT) in the evaluation of pulmonary infectious disease in the late period (>100 days) after allogeneic bone marrow transplantation (BMT). METHODS: Ninety-four matched CXR and CT examinations were performed for clinical suspicion of infectious lung disease. The time gap between CXR and CT was 48 h at maximum. The image pairs were correlated with the patients' clinical course and with the results of diagnostic bronchoalveolar lavage (BAL). An unremarkable clinical course over the subsequent seven days after imaging and/or negative microbiological culture served as the basis for excluding infectious lung disease. Positive microbiological culture and/or improvement of symptoms after antibiotic therapy were considered as evidence of infectious disease. RESULTS: The correlation with the clinical course and/or BAL revealed a significantly higher sensitivity, negative predictive value, and accuracy for CT than for CXR (89% versus 58%, P < 0.0001; 78% versus 47%, P < 0.0001; 90% versus 68%, P < 0.0001, respectively). CT was significantly more diagnostic in BAL verified fungal and bacterial infections (P < 0.05). CONCLUSION: CT is significantly superior to CXR in the evaluation of infectious pulmonary disease in the late phase after BMT. Therefore, an unremarkable CXR should be followed by a CT scan to reliably detect or to accurately exclude early pulmonary infection in these patients.
PURPOSE: To analyze the capabilities of chest roentgenogram (CXR) and computed tomography (CT) in the evaluation of pulmonary infectious disease in the late period (>100 days) after allogeneic bone marrow transplantation (BMT). METHODS: Ninety-four matched CXR and CT examinations were performed for clinical suspicion of infectious lung disease. The time gap between CXR and CT was 48 h at maximum. The image pairs were correlated with the patients' clinical course and with the results of diagnostic bronchoalveolar lavage (BAL). An unremarkable clinical course over the subsequent seven days after imaging and/or negative microbiological culture served as the basis for excluding infectious lung disease. Positive microbiological culture and/or improvement of symptoms after antibiotic therapy were considered as evidence of infectious disease. RESULTS: The correlation with the clinical course and/or BAL revealed a significantly higher sensitivity, negative predictive value, and accuracy for CT than for CXR (89% versus 58%, P < 0.0001; 78% versus 47%, P < 0.0001; 90% versus 68%, P < 0.0001, respectively). CT was significantly more diagnostic in BAL verified fungal and bacterial infections (P < 0.05). CONCLUSION: CT is significantly superior to CXR in the evaluation of infectious pulmonary disease in the late phase after BMT. Therefore, an unremarkable CXR should be followed by a CT scan to reliably detect or to accurately exclude early pulmonary infection in these patients.
Authors: Stephan Baumueller; Anna Winklehner; Christoph Karlo; Robert Goetti; Thomas Flohr; Erich W Russi; Thomas Frauenfelder; Hatem Alkadhi Journal: Eur Radiol Date: 2012-06-15 Impact factor: 5.315