Do specialists agree when interpreting contrast-enhanced endoscopic ultrasound?

Details

Contrast harmonic endoscopic ultrasound (CH-EUS) had recently entered clinical practice, but its reproducibility among endosonographers was unknown. The study aimed to assess interobserver agreement in CH-EUS.

Fifteen endosonographers, eight experienced and seven non-experienced, from 11 Italian endoscopic ultrasound centers evaluated 80 CH-EUS video cases: 40 solid pancreatic lesions, 20 pancreatic cystic lesions and 20 submucosal lesions. They assessed the degree of enhancement, the contrast agent’s distribution pattern and its washout. Agreement within and between the two observer groups was assessed using the Fleiss kappa statistic.

Overall agreement was moderate for uptake and fair for distribution pattern and washout. For solid pancreatic lesions, agreement was moderate for uptake and fair for pattern and washout. For cystic pancreatic lesions, agreement was uniformly moderate for uptake, slight for pattern and fair for washout. For submucosal tumors, agreement was substantial for uptake, slight for pattern and fair for washout. In most cases, non-experienced endosonographers showed agreement comparable to that of experienced endosonographers.

The researchers concluded that interobserver agreement for CH-EUS was satisfactory. Agreement ranged from slight to substantial and was fair in the majority of cases. The researchers suggested that inherent structural features of the lesions and technical differences between the variables assessed could explain the variation. Agreement outcomes were reproducible between experienced and non-experienced endosonographers.