Methods for the restoration of circulating blood volume, including the use of intravenous fluids, have been widely discussed over many years. There are no clear guidelines regarding the type of solutions, the total volume that should be transfused, or time schedules. Colloid solutions, usually hydroxyethyl starch compositions, are probably the most commonly used volume expanders in resuscitation, despite the lack of convincing trials and possible nephrotoxicity.In 2012, a task force of ESICM published a consensus statement on colloid use in critically ill adult patients. They stressed that infusion of an inappropriate volume may worsen the outcome of critically ill patients. Static parameters of cardiac filling volume, such as CVP or PCWP, commonly used in clinical practice, were found to be useless in the prediction of fluid responsiveness; volumetric or dynamic parameters, like global end diastolic volume (GEDV) or stroke volume variations (SVV), obtained by PICCO meters, seemed be much more appropriate. The dynamic fluid challenge test, which is transfusion of approx. 200 cc (or 3 mL kg-1) of any fluid over 5-10 min resulting in an increase of stroke volume, has also been recommended for the identification of those patients who may benefit from fluid resuscitation. The old passive leg raising test can also be used for this purpose.Despite prolonged discussion about fluid therapy in specific groups of critically ill patients, there is no convincing data to indicate the superiority of colloids over crystalloids. The choice of fluids is usually based on personal preference and hospital policy. Among crystalloid solutions, balanced preparations such as acetates, lactates, malates or citrates are recommended to avoid hyperchloraemia, a common side effect of saline infusion. There is no agreement regarding colloid solutions.The fluid transfusion regimen in critically ill patients should therefore be based on clinical assessment and patient responses.