Jeffery R. Galvin

FDG-PET imaging and the diagnosis of non-small cell lung cancer in a region of high histoplasmosis prevalence

STUDY OBJECTIVE Determine the sensitivity and specificity of [F-18]-fluorine-2-deoxy-D-glucose positron emission tomography (FDG-PET) in differentiating non-small cell lung cancer (NSCLC) from benign solitary pulmonary nodules (SPNs) in a region with a high endemic rate of histoplamosis. DESIGN Prospective, clinical study. SETTING University, tertiary referral hospital in the upper Mississippi River valley. PATIENTS Ninety patients with SPNs. INTERVENTIONS Independent interpretation of FDG-PET imaging, computed tomography and pathologic evaluation of the SPNs. MEASUREMENTS AND RESULTS To detect malignant SPNs, FDG-PET imaging had a sensitivity of 93%, a specificity of 40%, a positive predictive value (PPV) of 88% and a negative predictive value (NPV) of 55%. CONCLUSIONS In a region with a high prevalence of pulmonary fungal infection, FDG-PET is sensitive but has a low specificity and NPV for identifying NSCLC. In our study cohort, FDG-PET does not appear to reduce the need for SPN biopsies.

Evaluation of in vivo total and regional air content and distribution in primate lungs with high-resolution CT

RATIONALE AND OBJECTIVES The authors sought to determine whether gray-scale quantitative information from high-resolution computed tomography (CT) could reliably yield estimates of lung air content and help determine changes in air content with lung inflation and deflation. MATERIALS AND METHODS High-resolution CT images (n = 40) of lungs of two anesthetized monkeys were obtained after inflation with known air volumes. Percentage air content was calculated for each voxel, and lung volumes and patterns of air distribution were determined. RESULTS When corrected for pressure and temperature, high-resolution CT-based volumes correlated closely with inflation volumes (r = .99; standard error = 3.4%). Patterns of regional change in air content demonstrated known patterns of ventilation. CONCLUSION Although the high-spatial-frequency algorithm used in high-resolution CT enhances edges of structures and improves visualization of anatomic detail, gray-scale values from the same high-resolution CT data set remain a reliable index of regional lung attenuation.