This study reviewed the literature to directly evaluate the diagnostic performance of contrast-enhanced ultrasonography (CEUS) versus contrast-enhanced computed tomography (CECT) for assessing residual tumors of hepatocellular carcinoma treated with transarterial chemoembolization. PubMed, Embase, the Cochrane Library, and the China National Knowledge Infrastructure were searched through April 30, 2017. The pooled sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio, and summary receiver operating characteristic curve were calculated and compared to examine the diagnostic performance of CEUS versus CECT. A total of 11 studies, including 421 patients and 491 nodules were analyzed. The pooled diagnostic performances of CEUS versus CECT were as follows: (1) sensitivity (95% confidence interval), 0.97 (0.95-0.99) versus 0.72 (0.67-0.76); (2) specificity, 0.86 (0.74-0.94) versus 0.99 (0.95-1.00); (3) positive predictive value, 0.97 (0.95-0.99) versus 1.00 (0.98-1.00); (4) negative predictive value, 0.90 (0.83-0.95) versus 0.51 (0.44-0.58); (5) positive likelihood ratio, 7.79 (4.73-12.82) versus 12.50 (5.74-27.20); (6) negative likelihood ratio, 0.05 (0.03-0.09) versus 0.35 (0.26-0.48); (7) diagnostic odds ratio, 150.56 (57.03-397.49) versus 35.54 (14.89-84.83); and (8) area under the summary receiver operating characteristic curve, 0.9875 versus 0.9239. The sensitivity and negative predictive value of CEUS were significantly higher than those of CECT (both P < .001). The specificity and positive predictive value of CECT were significantly higher than those of CEUS (both P < .05). Contrast-enhanced US, with better sensitivity and negative predictive value versus CECT, was an effective method for exclusion of residual tumors after transarterial chemoembolization. Contrast-enhanced CT, with higher specificity than CEUS, is a valid approach for identifying residual tumors.