Pitfall in the detection of acute lesions of transient ischemic attack with fluid-inversion prepared diffusion weighted imaging

Objective To explore the limits of fluid-inversion prepared diffusion weighted imaging (FLIPD) in detection of acute cerebral ischemic lesions. Methods From January 2012 to March 2014, forty-nine patients (33 males, 16 females, age (55.6±12.3) years) clinically diagnosed as transient ischemic attack (TIA) were included. Patients underwent brain MRI (conventional diffusion weighted imaging (DWI) and FLIPD) within 3 d after the onset of TIA. The detection ability of MRI with the two sequences was compared, and the relative signal intensity (rSI) and apparent diffusion coefficient (ADC) of acute ischemic lesions based on two sequences were compared. Kappa test and two-sample t test were used to analyze the data. Results A total of 87 acute ischemic lesions were detected in 21 patients by conventional DWI, and 54 were detected in 19 patients by FLIPD (Kappa=0.916, P<0.05). The rSI of ischemic lesions on FLIPD was significantly lower than that on conventional DWI (1.37±0.22 vs 1.57±0.26; t=6.647, P<0.001). The ADC value of ischemic lesions on FLIPD was slightly lower than that on conventional DWI: (0.54 ±0.10) ×10-3 mm2/s vs (0.57±0.13)×10-3 mm2/s (t=2.120, P<0.05). The missed lesions on FLIPD were located in the white matter (n=18), cerebellum and brainstem (n=8), and the cortex (n=7). Conclusions A slight diffuse abnormality may be missed on FLIPD, so this method is not suitable for the detection of acute ischemic lesions. FLIPD technology still needs improvement. Key words: Ischemic attack, transient; Diffusion magnetic resonance imaging; Fluid attenuated inversion recovery sequence