ABSTRACT
PURPOSE: Purpose of this study was to determine if quantitative measures of CT attenuation and ADC values in combination with conventional imaging features can differentiate primary central nervous system lymphoma (PCNSL) and glioblastoma (GBM). MATERIALS AND METHODS: Twenty-six patients with histologically-proven GBM (14 men and 12 women; median age, 50 years; age range, 22 - 73 years) and 14 patients with PCNSL (11 men and 3 women; median age, 61 years; age range, 41 - 74 years) were enrolled. Maximum CT attenuation, minimum ADC, and lesion to normal parenchyma minimum ADC ratios were measured in solid tumor regions. Conventional imaging features were evaluated for the following: ill-defined margin, homogeneous enhancement pattern, degree of necrosis, extent of tumor involvement and multiplicity. The Mann-Whitney test was used to compare maximum CT attenuation and minimum ADC values for PCNSL and GBM. Fisher's exact test was used to evaluate relationships between pathologic diagnoses and imaging features. RESULTS: The CT attenuations were similar for PCNSL and GBM (37.84 +/- 6.90 HU versus 37.00 +/- 5.54 HU, p = 0.68), but minimum ADC and minimum ADC ratio were significant lower in PCNSL than in GBM (595.01 +/- 228.28 10(-6) mm2/s versus 736.52 +/- 162.05 10(-6) mm2/s; p = 0.028, 0.87 +/- 0.26 versus 1.14 +/- 0.29; p = 0.007). PCNSL showed greater homogeneous enhancement and smaller necrotic areas than GBM (p = 0.003 and p < 0.001, respectively) and was more likely to have multiple tumors than GBM (p = 0.039). When necrotic PCNSL (n = 4) and necrotic GBM (n = 24) were compared, minimum ADC and minimum ADC ratios were also significantly lower in PCNSL, but CT attenuation were not. CONCLUSION: Although CT attenuation does not provide valuable information, minimum ADC and minimum ADC ratio and some imaging features can aid the differentiation of PCNSL and GBM.