Crystalloids vs. colloids in fluid resuscitation: A systematic review

Objective: To systematically review the effects of isotonic crystalloids co mpared with colloids in fluid resuscitation. Data Sources: Computerized bibliographic search of published research and c itation review of relevant articles. Study Selection: All randomized clinical trials of adult patients requiring fluid resuscitation comparing isotonic crystalloids vs. colloids were incl uded. Pulmonary edema, mortality, and length of stay were evaluated. Indepe ndent review of 105 articles identified 17 relevant primary studies of 814 patients. Weighted kappa about article inclusion was high (0.76). Data Extraction: Data on population, interventions, outcomes, and methodolo gic quality of the studies were obtained by duplicate independent review wi th differences resolved by consensus. Weighted kappa on the validity assess ment was moderate (0.54). Data Synthesis: No difference was observed overall between crystalloid and colloid resuscitation with respect to mortality and pulmonary edema; howeve r, the power of the aggregated data was insufficient to detect small but po tentially clinically important differences. Subgroup analysis suggested a s tatistically significant difference in mortality in trauma in favor of crys talloid resuscitation (relative risk 0.39, 95% confidence intervals: 0.17 t o 0.89). Several methodologic issues are noteworthy regarding the primary s tudies, including lack of blinding (except in three studies). The type, dos e, and duration of fluid administration and outcomes measured were differen t across these trials. Conclusions: Overall, there is no apparent difference in pulmonary edema, m ortality, or length of stay between Isotonic crystalloid and colloid resusc itation. Crystalloid resuscitation is associated with a tower mortality in trauma patients. Methodologic limitations preclude any evidence based clini cal recommendations. Larger well-designed randomized trials are needed to a chieve sufficient power to detect potentially small differences in treatmen t effects if they truly exist.