Objective:

To evaluate the interobserver variability for different methods of tumor apparent diffusion coefficient (ADC) measurement in rectal cancer.

Methods:

A total of 37 patients with rectal cancer were analyzed retrospectively. They underwent MRI including diffusion weighted imaging (DWI) (b values: 0 and 1 000 s/mm2). The mean ADC values were measured by two readers using three region of interest (ROI) protocols: whole-volume

single slice and small solid samples. The values of ADC and standard deviation (SD) and interobserver variability were compared.

Results:

Intracl

ass correlation coefficient (ICC) for whole-volume measurement was excellent (0.87)

while for single slice and small solid sample measurements were moderate (ICC: 0.56 for single slice

0.58 for small solid sample). ADC values measured by small solid sample method were significantly lower than those by whole-volume (

P

0.001) and single slice (

P

0.001). The SDs calculated by whole-volume and single slice methods were significantly larger than that by small solid sample method (

P

0.001 and

P

=0.02

respectively).

Conclusion:

ADC values and interobserver variability are influenced by different measurement methods. There peatability of whole-volume method is the best.