Monaz: Hello, I'm Monaz Mehta, editor in the Cochrane Editorial and Methods department. The care of critically ill patients is highly complex, requiring treatments for their underlying conditions and for the consequences of those conditions, such as the loss of bodily fluids. A wide range of these treatments are assessed in Cochrane Reviews and one of these was updated in August 2018, looking at the effects of giving patients extra fluids. Lead author, Sharon Lewis from the Royal Lancaster Infirmary in the UK, tells us about the latest findings in this podcast.

Sharon: People who are critically ill may lose large amounts of fluid because of trauma, burns, infections, such as sepsis, or other serious conditions. They are often given additional fluids, usually intravenously, to try to counter this and two of the common types are crystalloids and colloids. Crystalloids are salt solutions, which are cheap, easy to use, and provide immediate resuscitation – but the small molecules in these solutions mean that they pass through the cells quickly and can cause oedema, or swelling. Colloids have larger molecules and may be more efficient at increasing fluid volume in the blood. They include starches, dextrans, gelatins, and naturally-occurring colloids, such as albumin or fresh frozen plasma or FFP. However, they are more expensive and there are concerns about side effects, including kidney failure, blood clotting disorders, and allergic reactions.
Therefore, we updated and extended the Cochrane Review of colloids versus crystalloids in critically ill people, to provide up-to-date evidence on effects on death, blood transfusion or renal replacement therapy, and adverse events, in particular, allergic reactions, itching, or rashes.
We focused on critically ill people who needed fluid volume replacement in hospital or in an emergency out-of-hospital setting, but didn’t include studies of new born babies, women who had caesarean sections, or people scheduled for any type of surgery.
We found 65 randomised and 4 quasi-randomised trials, with a total of just over 30,000 participants. Four types of colloids had been tested against crystalloids: starches in 28 studies, dextrans in 20, gelatins in 7, and albumin or FFP in the other 22. 
Evaluating each of these colloids, we found little or no difference between starches and crystalloids on deaths within 30 or 90 days, or by the final time point reported in each study. However, starches probably slightly increase the need for blood transfusion and for renal replacement therapy. Fewer participants given crystalloids reported itching or rashes but too few studies reported whether starches caused adverse events for us to be confident about which is better or worse.
We judged the evidence on starches to be of moderate certainty and felt the same about the evidence for dextrans, and albumin or FFP. These colloids also probably make little or no difference to mortality compared to crystalloids and the finding was similar for gelatins versus crystalloids, but, because we had fewer studies for this comparison, we judged the evidence to be of low certainty. We also cannot be certain about whether there are any differences between these types of colloid and crystalloids on the need for a blood transfusion or renal replacement therapy, or on adverse events.
This is the sixth update of this review, and there remains some hope that the certainty of the evidence will increase in future updates, because we found some studies that could not be included at the moment and there are at least three ongoing studies, which should become available in the coming years.

Monaz: If you would like to read the full results of the latest update, and watch for the next one, you can find it with a search for 'colloids versus crystalloids' at Cochrane Library dot com.