Arterial spin labelling and diffusion-weighted magnetic resonance imaging in differentiation of recurrent head and neck cancer from post-radiation changes.

OBJECTIVE: To assess arterial spin labelling and diffusion-weighted imaging in the differentiation of recurrent head and neck cancer from post-radiation changes.
METHODS: A retrospective study was conducted of 47 patients with head and neck cancer, treated with radiotherapy, who underwent magnetic resonance arterial spin labelling and diffusion-weighted magnetic resonance imaging. Tumour blood flow and apparent diffusion co-efficient of the lesion were calculated.
RESULTS: There was significant difference (p = 0.001) in tumour blood flow between patients with recurrent head and neck cancer (n = 31) (47.37 ± 16.3 ml/100 g/minute) and those with post-radiation changes (n = 16) (18.80 ± 2.9 ml/100 g/minute). The thresholds of tumour blood flow and apparent diffusion co-efficient used for differentiating recurrence from post-radiation changes were more than 24.0 ml/100 g/minute and 1.21 × 10-3 mm2/second or less, with area under the curve values of 0.94 and 0.90, and accuracy rates of 88.2 per cent and 88.2 per cent, respectively. The combined tumour blood flow and apparent diffusion co-efficient values used for differentiating recurrence from post-radiation changes had an area under the curve of 0.96 and an accuracy of 90.2 per cent.
CONCLUSION: Combined tumour blood flow and apparent diffusion co-efficient can differentiate recurrence from post-radiation changes.