摘要
Background: Differentiating cryptogenic organizing pneumonia (COP) from community-acquired pneumonia (CAP) can be difficult in some cases. Objective: To clarify the clinical utility of procalcitonin (PCT) levels for differentiating between COP and CAP. Methods: Blood PCT levels, leukocyte count, C-reactive protein concentration, number of segments involved on computed tomography (CT) images, and pneumonia severity assessment scale were retrospectively investigated from clinical charts and chest CT images of COP and CAP patients who were admitted to our hospital from 2012 to 2014. The clinical characteristics of COP patients were compared to those of CAP patients for whom causative organisms were not detected. Results: There were 16 COP and 94 CAP patients for whom causative organisms were not detected. Blood PCT levels of all COP patients were less than 0.16 ng/dL, and significantly lower than those of CAP patients (p = 0.0004), while the number of involved segments was significantly higher than that of CAP patients (p = 0.0001). Blood PCT levels and the number of involved segments remained independently significant for differentiating between COP and CAP by multivariate analysis. Receiver operating characteristics curve analysis revealed that 7 was the best cut-off number for involved segments to differentiate between COP and CAP patients with low PCT levels (sensitivity 85.7%, specificity 94.7%). Conclusion: A combination of PCT levels and number of involved segments on CT images is useful for differentiation between COP and CAP.
Background: Differentiating cryptogenic organizing pneumonia (COP) from community-acquired pneumonia (CAP) can be difficult in some cases. Objective: To clarify the clinical utility of procalcitonin (PCT) levels for differentiating between COP and CAP. Methods: Blood PCT levels, leukocyte count, C-reactive protein concentration, number of segments involved on computed tomography (CT) images, and pneumonia severity assessment scale were retrospectively investigated from clinical charts and chest CT images of COP and CAP patients who were admitted to our hospital from 2012 to 2014. The clinical characteristics of COP patients were compared to those of CAP patients for whom causative organisms were not detected. Results: There were 16 COP and 94 CAP patients for whom causative organisms were not detected. Blood PCT levels of all COP patients were less than 0.16 ng/dL, and significantly lower than those of CAP patients (p = 0.0004), while the number of involved segments was significantly higher than that of CAP patients (p = 0.0001). Blood PCT levels and the number of involved segments remained independently significant for differentiating between COP and CAP by multivariate analysis. Receiver operating characteristics curve analysis revealed that 7 was the best cut-off number for involved segments to differentiate between COP and CAP patients with low PCT levels (sensitivity 85.7%, specificity 94.7%). Conclusion: A combination of PCT levels and number of involved segments on CT images is useful for differentiation between COP and CAP.
