Novel quantitative perfusion analysis with contrast-enhanced harmonic EUS for differentiation of autoimmune pancreatitis from pancreatic carcinoma

Abstract

Objective. Autoimmune pancreatitis (AIP) is often misdiagnosed as pancreatic carcinoma (PC) despite recent advances in imaging tests. The aim of the study was to evaluate whether the quantitative perfusion analysis using software “Time intensity curve” with contrast-enhanced harmonic EUS (CH-EUS) facilitate the differentiation of AIP from PC. Methods. Consecutive patients with focal AIP and pancreatic carcinoma who underwent CH-EUS from January 2009 to September 2010 were analyzed. CH-EUS was performed with intravenous administration of an ultrasonographic contrast (Sonazoid) and electronic radial echoendoscope. The graph of time intensity curve (TIC) for pancreatic mass was generated to depict the changes in signal intensity over time within the region of interest (ROI). ROI was placed to cover an area with a pancreatic mass lesion. Based on the analysis of TIC, base intensity before injection (BI), peak intensity (PI), time to peak, and maximum intensity gain (MIG: PI-BI) were calculated. Results. Eight patients with focal AIP and twenty-two patients with PC were evaluated by TIC. PI and MIG of mass lesion of AIP were significantly higher than that of PC (21.4 dB vs. 9.6 dB, 17.5 vs. 6.6). Receiver operating characteristics analysis yielded an optimal MIG cutoff value of 12.5 with high sensitivity and specificity. Conclusion. Pancreatic mass lesions of AIP and PC exhibited markedly different patterns with the TIC. This novel diagnostic modality using TIC generated by CH-EUS might offer an opportunity to improve accuracy in the differential diagnosis between pancreatic mass lesion of AIP and PC.

Key Words::

• autoimmune pancreatitis
• contrast-enhanced harmonic EUS
• pancreatic carcinoma

Acknowledgement

This study was approved with the Ethical Committee of our institution on 6 February 2012, and registration number is 23–251, 6712.

Declaration of interest: The authors have no conflict of interest and no financial or personal relationships that could influence or be perceived to influence the work presented in this article.