The diagnostic performance of contrast-enhanced ultrasound versus contrast-enhanced computed tomography for pancreatic carcinoma: a systematic review and meta-analysis

Xiaoyi Yan,Ke Lv,Mengsu Xiao,Li Tan,Yang Gui,Jing Zhang,Xueqi Chen,Wanying Jia,Jinglin Li
DOI: https://doi.org/10.21037/tcr-22-601
Abstract:Background: Pancreatic carcinoma is a highly fatal disease, and early diagnosis is of vital importance. This meta-analysis aimed to compare the diagnostic performances of contrast-enhanced ultrasonography (CEUS) against contrast-enhanced computed tomography (CECT) for pancreatic carcinoma, using pathological results or alternative imaging modality as the gold standard. Methods: A thorough search was conducted in PubMed, EMBASE, Web of Science and Cochrane Library databases. Two investigators selected the studies and extracted the data independently. A bivariate mixed-effects regression model was used to calculate the pooled data. Subgroup analysis and meta-regression were performed to explore the causes of heterogeneity. Results: A total of 1,227 records were identified, of which 7 articles with 588 patients were assessed for eligibility. The overall sensitivity, specificity of CEUS and CECT with their 95% confidential intervals (95% CI) were 0.91 (0.85-0.94) and 0.88 (0.81-0.92), 0.83 (0.70-0.91) and 0.87 (0.73-0.94), respectively. The area under curve (AUC) of CEUS and CECT were 0.94 and 0.93. Subgroup analysis showed CEUS may be good at diagnosing lesions with diameters less than 2 cm. Tumor features, region and study type were the main causes of heterogeneity. Conclusions: CEUS has a satisfying diagnostic performance for pancreatic carcinoma and it has high sensitivity for small pancreatic carcinomas (≤2 cm); besides, it performs well in discriminating pancreatic cancer from chronic pancreatitis. Therefore, CEUS can be a useful supplement to CECT.
What problem does this paper attempt to address?