Fifteen studies were included (1,722 patients): four RCTs (randomised controlled trials, 272 patients), three prospective cohort studies (343 patients), four retrospective cohort studies (563 patients), one prospective case-control study (129 patients) and three case series (415 patients).
Type of fluid: One RCT (41 patients) found that Ringer lactate with hydroxyethyl starch significantly reduced intra-abdominal pressure on days two to seven when compared with Ringer lactate alone. Another RCT (40 patients) reported significant reductions from baseline in systemic inflammation in patients who received Ringer lactate but no such reduction in the normal saline comparison group (between-group comparisons were not presented). Neither study reported statistically significant differences between groups for organ failure, length of hospital stay and mortality.
Non-aggressive versus aggressive fluid resuscitation: There was moderate-quality evidence that non-aggressive fluid therapy resulted in lower organ dysfunction (RR 0.69, 95% CI 0.54 to 0.88; one RCT) and lower mortality (RR 0.40, 95% CI 0.22 to 0.72; two RCTs) compared with aggressive fluid resuscitation. The quality of the evidence relating to the other reported outcomes was mostly very low.
The evidence for goal-directed fluid therapy was sparse with low numbers of events.