BACKGROUND: Although targeted by radiotherapy, recurrence in glioblastoma occurs mainly periresectionally owing to tumor infiltration. An increase in the apparent diffusion coefficient (ADC) has been shown in the large high-T2 area on magnetic resonance imaging posttreatment; however, until now ADC has not been investigated directly in the more relevant periresectional area. METHODS: Histogram analysis was used to assess periresectional ADC values in patients with glioblastoma postradiotherapy versus preradiotherapy. Periresectional ADC values starting at 0-5 mm in 5-mm increments up to 20-25 mm were extracted and compared using 2-way repeated-measurements analysis of variance. RESULTS: Mean ADC values directly adjacent to the resection area (0-5 mm) were significantly higher postradiotherapy compared with preradiotherapy (P = .017). ADC values in the 0- to 5-mm region were also higher than those in 5- to 10-, 10- to 15-, and 15- to 20-mm regions (P < .05). Regional standard deviations in ADC values were higher postradiotherapy compared with preradiotherapy for the 0- to 5-mm region up to the 15- to 20-mm region, inclusive (P < .05); however, Cox regression analysis showed no survival benefits from the increased ADC in the 0- to 5-mm region postradiotherapy. CONCLUSIONS: Increased ADC values, representing a decrease in infiltrative tumor load, were demonstrated in a limited direct periresectional area. This finding adds to previous studies evaluating ADC response in the larger high-T2 area in relation to survival.
BACKGROUND: Although targeted by radiotherapy, recurrence in glioblastoma occurs mainly periresectionally owing to tumor infiltration. An increase in the apparent diffusion coefficient (ADC) has been shown in the large high-T2 area on magnetic resonance imaging posttreatment; however, until now ADC has not been investigated directly in the more relevant periresectional area. METHODS: Histogram analysis was used to assess periresectional ADC values in patients with glioblastoma postradiotherapy versus preradiotherapy. Periresectional ADC values starting at 0-5 mm in 5-mm increments up to 20-25 mm were extracted and compared using 2-way repeated-measurements analysis of variance. RESULTS: Mean ADC values directly adjacent to the resection area (0-5 mm) were significantly higher postradiotherapy compared with preradiotherapy (P = .017). ADC values in the 0- to 5-mm region were also higher than those in 5- to 10-, 10- to 15-, and 15- to 20-mm regions (P < .05). Regional standard deviations in ADC values were higher postradiotherapy compared with preradiotherapy for the 0- to 5-mm region up to the 15- to 20-mm region, inclusive (P < .05); however, Cox regression analysis showed no survival benefits from the increased ADC in the 0- to 5-mm region postradiotherapy. CONCLUSIONS: Increased ADC values, representing a decrease in infiltrative tumor load, were demonstrated in a limited direct periresectional area. This finding adds to previous studies evaluating ADC response in the larger high-T2 area in relation to survival.