A total of 17 studies were included in the analyses: nine RCTs (only one described the randomization method used); five controlled trials; and three self-controlled studies. No study reported allocation concealment, patient blinding or investigator blinding. Seven studies (five RCTs and two self-controlled studies) reported blinding of outcome assessors.
For clinical images there was no significant difference in mean contrast enhancement, with or without saline flush, for the liver (mean difference −0.4, 95% CI −2.4 to 1.7; two studies) or the abdominal aorta (mean difference 11, 95% CI −5.3 to 27; two studies). Saline flush significantly improved contrast enhancement for the portal vein (mean difference 7.4, 95% CI 1.3 to 14; three studies).
For time-density analysis, saline flush significantly improved contrast enhancement in all anatomical areas studies: liver (mean difference 8.2, 95% CI 4.2 to 12; two studies); portal vein (mean difference 28, 95% CI 16 to 40; two studies) and abdominal aorta (mean difference 15, 95% CI, 0.3 to 30; three studies).
The results of between-study heterogeneity assessment were not reported.