Colloid Solutions For Fluid Resuscitation In Sepsis: Systematic Review Of Randomized Controlled Trials

Objective: We aim to assess the effects on mortality and safety of different colloid solutions in patients with sepsis requiring volume replacement, by examining direct comparisons of colloid solutions. Methods: We searched the Cochrane Central Register of Controlled Trials (CENTRAL) on The Cochrane Library, MEDLINE (1966 to present), EMBASE (1974 to present), and the China Biological Medicine Database (1978 to present), VIP Chinese Journals Database (1989 to present), CNKI China National Knowledge Infrastructure Whole Article Database (1994 to present) by computer. The search was last updated in June 2010. Randomized clinical trials which comparing different colloids in septic patients who needing fluid resuscitation were selected. Studies were separately selected by three unblended review authors. Two of them also independently extracted data. Results: 17 randomized clinical trials with a total 1281 participants met the inclusion criteria. Mortality was obtained in all trials. For intervention of albumin versus hydroxyethyl starch (HES) 10 trials (n = 413) reported mortality, the RR (relative risk) of death was 0.98 [0.74 to 1.30]. Intervention of albumin versus gelatin, two trials (n = 100) reported mortality. The RR (relative risk) of death was 2.4[95% CI 0.31 to 18.35].intervention of gelatin versus HES four trials (n = 205) reported mortality, the RR (relative risk) of death was 1.02[95% CI 0.79 to 1.32]. Intervention of HES versus dextran two trials (n=463) reported mortality, the RR of death was 1.38[95% CI 0.8to 6.78]. Intervention of gelatin versus dextran one trial (n = 112) reported mortality, RR (relative risk) of death was not estimable. For albumin and HES versus Dextran no trial was included. Four trials of intervention of albumin versus HES recorded the change of severity score. Conclusion: 1) There is no evidence that one colloid solution is more effective and safer than another for the fluid resuscitation in sepsis. 2) The severity score is improved in HES, but the confidence intervals are wide. 3) More clinical trials are needed to evaluate the effect and the safety of different colloids in sepsis, especially about children.