Analysis of the layering pattern of the apparent diffusion coefficient (ADC) for differentiation of radiation necrosis from tumour progression

Objectives

To evaluate the added value of diffusion-weighted imaging (DWI) to perfusion-weighted imaging (PWI) for differentiating tumour progression from radiation necrosis.

Methods

Sixteen consecutive patients who underwent removal of a metastatic brain tumour that increased in size after stereotactic radiosurgery were retrospectively reviewed. The layering of the ADC was categorised into three patterns. ADC values were measured on each layer, and the maximum rCBV was measured. rCBV and the layering pattern of the ADC of radiation necrosis and tumour progression were compared.

Results

Nine cases of radiation necrosis and seven cases of tumour progression were pathologically confirmed. Radiation necrosis (88.9 % vs. 14.3 %) showed a three-layer pattern of ADC with a middle layer of minimum ADC more frequently. If rCBV larger than 2.6 was used to differentiate radiation necrosis and tumour progression, the sensitivity was 100 % but specificity was 56 %. If the lesions with the three-layer pattern of ADC with moderately increased rCBV (2.6–4.1) were excluded from tumour progression, the sensitivity and specificity increased to 100 %.

Conclusions

The three-layer pattern of ADC shows high specificity in diagnosing radiation necrosis; therefore, combined analysis of the ADC pattern with rCBV may have added value in the correct differentiation of tumour progression from radiation necrosis.

Key Points

?MRI follow-up often reveals increasing size of enhancing lesions after stereotactic radiosurgery ?Inflammation around tumour necrosis increases regional cerebral blood volume (rCBV), mimicking progression ?A three-layer apparent diffusion coefficient (ADC) pattern on diffusion-weighted MRI suggests radiation necrosis. ?The combined use of rCBV and DW MRI enables accurate differentiation.