Clinical application of fluid attenuated inversion recovery in apparent diffusion coefficient quantitative measurements of ischemic brain infarction / 中华放射学杂志

ABSTRACT

Objective To quantitatively evaluate the influence of fluid attenuated inversion prepared recovery (FLAIR) on apparent diffusion coefficient (ADC) and its clinical application value. Methods The data of DWI and FLAIR-DWI of 139 stroke were retrospectively reviewed. Paired t-test was used to analyze DWI (ADCCON ) and FLAIR-DWI (ADCFLAIR) values at varying time points from hyperacute to chronic stage. All of the lesions were further divided into cortex involved infarction and subcortical infarction. The ADCCON and ADCFLAIR values in the lesion sides and the contralateral sides were compared separately. Results The mean ADCCON values for lesions less than 6 hours, 7--12 hours, 13--24 hours, within 2 days, 3-4 days, 5-7 days and 8--14 days were not significantly different from those of the ADCFLAIR values(P >0.05) [ADCCON were (0.55±0.07), (0.50±0.09), (0.50±0. 13), (0.50 ± 0. 13), (0.62 ± 0. 14), ( 0. 60 ± 0. 12), (0. 72 ± 0. 20) × 10-3 mm2/s; ADCFLAIR were ( 0. 53 ± 0. 09 ), (0.49±0.06),(0.49±0.10),(0.48±0.08),(0.58±0. 14), (0.60±0.09),(0.73±0.15) × 10-3 mm2/s]. Lesions of 15 to 30 days [ (0. 95±0. 21 ) × 10-3 mm2/s and ( 1.02±0. 27) × 10-3 mm2/s for ADCFLAIR and ADCCON ] and the chronic stage ( >31 days) [ ADCFLAIR and ADCCON were (1.10 ± 0. 30) × 10-3 mm2/s and (1.36±0. 41 ) × 10-3 mm2/s respectively], had a significantly lower ADCFLAIR than those of the ADCCON (P <0. 01 ). For patients with a symptom duration of less than 14 days, the mean ADCFLAIR values of the cortex involved and subcortical lesions were all not significantly different from the mean ADCCON (P > 0. 05 ) [ ADCCON were ( 0. 55 ± 0. 16 ), ( 0. 61 ± 0. 14 ) × 10-3 mm2/s ; ADCFLAIR were (0.53±0. 14), (0.60±0. 13) × 10-3 mm2/s]. For patients with a symptom duration of longer than 14 days, the mean ADCFLAIR values of the cortex involved and subeortical lesions were all significantly lower than those of the mean ADCCON values [ ( 1.16±0. 36) × 10-3 mm2/s vs. ( 1.35±0. 48) × 10-3 mm2/s for cortex involved lesions and (0. 97±0. 19) × 10-3 mm2/s vs. ( 1.15±0. 33) × 10-3 mm2/s for subcortical lesions ] (P < 0. 01 ). The ADC values of the normal contralateral sides were significantly decreased after the fluid inversion prepared pulse was conducted [ ADCFLAIR, ( 0. 76 ± 0. 05 ) × 10-3 mm2/s and ADCCON, (0. 82 ± 0. 11 ) × 10-3 mm2/s ] ( p < 0. 01 ). Conclusions The FLAIR significantly decrease the absolute ADC values of the ischemic lesions 14 days later after the stroke onset, which may be helpful in determining individual lesion age. Meanwhile, the application of FLAIR can have a more accurate relative ADC value by reducing the free fluid partial volume effect of the normal contralateral side, and hence enhance the ability of detecting the subtle ischemic pathophysiological changes.