Objective: The purpose of our study was to assess the reliability and usefulness of parallel imaging for apparent diffusion coefficient (ADC) measurement of abdominal organs and lesions.
Materials and methods: Single-shot spin-echo echo-planar diffusion-weighted MRI (TE = 66, b = 0, 600 s/mm2) was performed in phantom and clinical studies. The b value was set to minimize the effects of perfusion in tissue and to maintain signal-to-noise ratio. Bottle phantoms were scanned with and without parallel imaging and with various parallel imaging factors and at various positions to evaluate the effects of parallel imaging on ADCs. In 200 consecutive clinical patients (122 men and 78 women: mean age, 61.9 years), ADCs were calculated for liver (four segments), spleen, pancreas (head, body, tail), gallbladder, renal parenchyma, and back muscle, and then compared to evaluate the reliability of clinical ADC measurements with parallel imaging. ADCs were also calculated for diffuse diseases and focal lesions (94 malignant and 93 benign) of abdominal organs to evaluate the clinical usefulness of ADC.
Results: Location-dependent changes in water ADCs were minimal with parallel imaging factors first of 3, then of 4, and were small except for measurements at the image periphery. Acetone ADCs were saturated at 4.00 x 10(-3) mm2/s. Degraded image quality prevented ADC measurement of the left hepatic lobe and pancreas in 7-18 patients. There was no significant difference among ADCs of four liver segments (1.50 +/- 0.24 [SD] x 10(-3) mm2/s - 1.56 +/- 0.31 x 10(-3) mm2/s) and between ADCs of the right and left kidneys (2.65 +/- 0.30 x 10(-3) mm2/s, 2.59 +/- 0.33 x 10(-3) mm2/s). ADC of the pancreas tail (1.65 +/- 0.37 x 10(-3) mm2/s) was significantly lower than those of the head (1.81 +/- 0.40 x 10(-3) mm2/s) and body (1.81 +/- 0.41 x 10(-3) mm2/s) (p < 0.005). Renal ADCs were significantly lower in patients with renal failure (right: 2.15 +/- 0.30 x 10(-3) mm2/s; left: 2.11 +/- 0.25 x 10(-3) mm2/s) than in those without disease (right: 2.67 +/- 0.29 x 10(-3) mm2/s; left: 2.60 +/- 0.32 x 10(-3) mm2/s) (p < 0.005). ADC of pancreatic cancer was significantly higher than that of healthy pancreas (p < 0.05). ADC of renal angiomyolipoma was significantly lower than those of renal cell carcinoma and healthy renal parenchyma (p < 0.0005).
Conclusion: Clinical ADC measurements of abdominal organs and lesions using parallel imaging appear to be reliable and useful, and the effect of parallel imaging on calculated values is considered to be minimal.