Cerebellar tumors in pediatric patients: can they be differentiated using diffusion-weighted imaging?

    loading  Checking for direct PDF access through Ovid

Abstract

BACKGROUND

The differential diagnosis of cerebellar neoplasms in children and adolescents can be challenging, but these tumors show considerable differences in cell attenuation and integrity. It is possible that these differences can be detected indirectly by measuring the water diffusivity of tumor tissue.

OBJECTIVE

To determine the apparent diffusion coefficient (ADC) values of tissue water in various pediatric cerebellar tumors using diffusion-weighted imaging (DWI).

DESIGN AND INTERVENTION

This retrospective review of patient records was performed at two centers in the US and included 32 patients (mean age 9 years, range 6 weeks to 23 years) with cerebellar neoplasms. The tumors were histologically classified as follows: juvenile pilocytic astrocytoma (JPA; n = 17), medulloblastoma (n = 8), ependymoma (n = 5), and atypical teratoid/rhabdoid tumor (AT/RT; n = 2). All patients underwent conventional MRI and DWI. The scans were evaluated using a region-of-interest (ROI)-based approach, with either one or three 50-100 mm2 regions (1-ROI and 3-ROI method, respectively) located on homogeneous enhancing solid areas of the tumor. Control ADC values were obtained from normal brain tissue (cerebellum [1-ROI method], or cerebellum and bilateral centrum semiovale [3-ROI method]). The ratios of the average tumor ADC to the average control ADC were also determined.

OUTCOME MEASURES

The absolute ADC values and the ADC ratios were the outcome measures.

RESULTS

Normal brain ADC values were significantly different from the ADC values of cerebellar tumors (P <0.05 for all tumor types). JPAs and medulloblastomas could be differentiated using both absolute ADC values and ADC ratios. The mean ADC value using the 3-ROI method was 1.65 ± 0.27 × 10−3 mm2/s for JPAs and 0.66 ± 0.15 × 10−3 mm2/s for medulloblastomas (P <0.0001). A comparable result was obtained using the 1-ROI approach. There was no overlap in individual ADC values, with an ADC range of 1.24-2.09 × 10−3 mm2/s for JPAs and 0.48-0.93 × 10−3 mm2/s for medulloblastomas (1-ROI and 3-ROI methods). The ADC ratios of JPAs and medulloblastomas were also significantly different (P <0.0001 for both the 1-ROI and the 3-ROI technique), and there was no overlap of ADC ratio ranges for these two tumor types (ADC ratio ranges obtained with both methods: 1.62-2.99 for JPAs and 0.66-1.10 for medulloblastomas). All measures obtained for ependymomas were significantly different from those obtained for JPAs (P <0.05) or medulloblastomas (P <0.001), with an ADC range for ependymomas of 0.97-1.29 × 10−3 mm2/s (1-ROI and 3-ROI methods). The measures obtained for AT/RT were within the range of data for medulloblastomas. ADC values of above 1.40 × 10−3 mm2/s were 100% specific for JPAs, whereas values below 0.90 × 10−3 mm2/s were 100% specific for a diagnosis of medulloblastoma or AT/RT. Most ependymomas had ADCs between 1.00 and 1.30 × 10−3 mm2/s.

CONCLUSION

Cerebellar tumors in children and adolescents can be differentiated by measuring the ADC values of enhancing solid tumor tissue.

Related Topics

    loading  Loading Related Articles