Magnetic resonance imaging of pulmonary infection in immunocompromised children: comparison with multidetector computed tomography.

BACKGROUND

Computed tomography (CT) is commonly used to detect pulmonary infection in immunocompromised children.

OBJECTIVE

To compare MRI and multidetector CT findings of pulmonary abnormalities in immunocompromised children.

MATERIALS AND METHODS

Seventeen neutropaenic children (6 girls; ages 2-18 years) were included. Non-contrast-enhanced CT was performed with a 64-detector CT scanner. Axial and coronal non-enhanced thoracic MRI was performed using a 1.5-T scanner within 24 h of the CT examination (true fast imaging with steady-state free precession, fat-saturated T2-weighted turbo spin echo with motion correction, T2-weighted half-Fourier single-shot turbo spin echo [HASTE], fat-saturated T1-weighted spoiled gradient echo). Pulmonary abnormalities (nodules, consolidations, ground glass opacities, atelectasis, pleural effusion and lymph nodes) were evaluated and compared among MRI sequences and between MRI and CT. The relationship between MRI sequences and nodule sizes was examined by chi- square test.

RESULTS

Of 256 CT lesions, 207 (81%, 95% confidence interval [CI] 76-85%) were detected at MRI. Of 202 CT-detected nodules, 157 (78%, 95% CI 71-83%) were seen at motion-corrected MRI. Of the 1-5-mm nodules, 69% were detected by motion-corrected T2-weighted MRI and 38% by HASTE MRI.

CONCLUSION

Sensitivity of MRI (both axial fat-saturated T2-weighted turbo spin echo with variable phase encoding directions (BLADE) images and HASTE sequences) to detect pulmonary abnormalities is promising.