The use of Intravenous fluid therapy and the controversies around them have been a theme of the blog from the very beginning. From the early days when evidence increasingly pointed to the harms of starch based solutions1–5 through to one of our latest podcasts on the use of balanced solutions in children6, the use of intravenous therapies has been a core of interest, uncertainty and dogmalysis.
Let’s get a few really important points of our chest right now.
- Fluids are drugs
- Like pretty much every other drug these are drugs that can cure but they can also harm and they may even kill
- The physiological response to fluids is dependent on the patient, the pathology and their physiology.
- IV fluids are probably the most commonly prescribed therapy in hospital medicine
- The evidence base is disappointingly poor, but is improving rapidly with the advent of large, well conducted, pragmatic, clinician led trials.
- There is nothing normal about ‘Normal Saline’
- The FEAST trial7 was a game changer for us. It really challenged our view on fluids management and should be recognised as a turning point in the way that we think about sepsis management. It was influential not just for its results, but also for the debate and controversies that arose from its findings8–10.
The bottom line is that it’s a complex area, and the research field is moving pretty quickly which is why it’s helpful to periodically stop and digest a really well written review article to embed some core knowledge before embarking on the next stages of critical appraisal and review of individual trials over the next few years. This week we have a highly accessible article to review that should help us do exactly that.
That is why we are recommending that you read the paper by Finfer, Myburgh and Bellomo in this months Nature Reviews Nephrology11. The abstract is below, but as always, and especially so in this case, we strongly advocate that you read the full article.
I would describe this as a narrative review. It represents the expert opinion of the authors based on their clinical practice and their own research which is also heavily cited within the paper. This fact is important when ‘reviewing reviews’ where there is the significant potential for bias based on the authors personal beliefs and experiences.
This may concern you, and it should in all cases inject a degree of scepticism when you read a paper constructed in this way as opinions based on one’s own research are likely to take a certain viewpoint.
This does not mean that the paper is fatally flawed or wrong, it just means that there may be alternative perspectives that may not be represented here, and although my interpretation of the evidence in this review is very closely aligned to the authors it is fair to say that the views here are not universally held. For example I was recently at a seminar where the use of Starch solutions was advocated for septic patients, a view which I find very difficult to understand in light of the evidence described in the paper, but there you go.
The bottom line here is that although we believe that this paper represents a state of the art review of the evidence we also think that you should engage your brain and, as always, maintain your sense of scepticism.
What are the main lessons to take away from this paper?
These are my headline notes after reading the paper. I’m interested to see if others took away the same messages.
- The evidence base for many if not most of IV fluid therapy is terrible. Just read the comments on how certain solutions came into practice on the basis of trials with fewer than 250 patients and 48-hour end points. How can this possibly be for such an important and widely prescribed therapy.
- The approach to fluid therapy that I grew up with (fill them up and then add some more) may be harmful in critically ill patients.
- Fluid therapy should be tailored to which phase of shock the patients is in. The model of a 4-phase description of shock phases (Salvage – Optimisation – Stabilisation – De-escalation) should be matched to a different strategy for fluid management.
- Fluid therapy will also be influenced by the type of shock and the relative influence of each type which may occur in the same patient i.e. combinations of Hypovolaemia, Distributive, Obstructive, Cardiogenic shock may occur in the same patient. Echo may help differentiate these shock states and determine the relative contributions in complex patients12.
- We are increasingly understanding the importance of the role of the glycocalyx in fluid distribution and capillary permeability. This further understanding is changing the way that we think about oncotic pressure in the capillary circulation (and thus the role of crystalloids/colloids in critical illness). This is revising our understanding of the Starling theory of capillary/tissue perfusion and questioning the theoretical advantage of colloids in critical illness (which is in keeping with their lack of benefit in pragmatic clinical trials).
- It’s worthwhile reviewing and revising the various different components and subsequent physiological effects of the wide range of fluids that are commercially available. It’s also worth noting that there is (again) a paucity of data to reliably support the administration of one fluid over another in the vast majority of clinical syndromes associated with critical illness.
- There is a consistent disparity between the observed effects of different fluids in clinical trials as compared to that predicted from laboratory or theoretical models. To me this suggests that we have not really understood the physiological processes within critical illness and thus that it is dangerous (word chosen carefully) to make assumptions about physiological effects on the basis of standard models of physiology in health and disease. Any view might be to conclude that there is still much to learn about the physiological response to disease and our subsequent interventions.
- Hydroxy-Ethyl Starch solutions have been shown to be harmful and should not be used in the critically ill patient. They cause increases in mortality and the need for renal replacement therapy13,14,15,16.
- Gelatin based solutions are significantly under-investigated but are likely to have similar effects to HES solutions.
- In patients with AKI or at risk of AKI the current paradigm is to maintain volume and to promote diuresis where possible. However, this approach of ensuring that that the patient avoids hypovolaemia and that in many cases they risk fluid overload. Recent studies suggest that a more conservative approach to fluid therapy may reduce the incidence of AKI in critically ill patients. There is insufficient evidence to currently advocate restrictive or liberal fluid approaches with respect to AKI. There is clearly more work to be done.
- Human Albumin is similar in performance to the majority of critically ill patients but may have benefits in some subgroups such as sepsis and bacterial peritonitis17,18. It should not be used in traumatic brain injury.
- It is likely, but not yet definitive that buffered solutions (e.g. Ringer’s, Plasmalyte) are advantageous as compared to Saline. Evidence comes from large open label randomised controlled trials, but these are at risk of bias and we await further information from large RCTs which are currently in progress.
- Determining outcomes for trials of fluids in critically ill patients is complex. Mortality is clearly important but we should also look at general health assessments as well as proxy markers such as renal function and the need for renal replacement therapy.
Further reading (and listening)
There is already a lot out there in the #FOAMed world and beyond. These are resources I’ve found useful.
The revised Starling curve and the glycocalyx Tom Woodcock20
The Revised Starling Principle: Implications for Rational Fluid Therapy Jon-Emile Kenny 21
Is it really all about the glycocalyx? SMACC Talk from Jeremy Cohen22
There is clearly much more in the paper that is worthy of a read and I strongly recommend you do. I hope that it helps you understand where we are in terms of the knowledge around the administration of IV fluids to critically ill patients, but I also hope that this paper makes you uncomfortable. It clearly sets out the uncertainty that underpins our practice for the most challenging and seriously unwell patients in our care. Most of our interventions are based on custom and practice and/or courageous assumptions about how fluids work in illness. There is hope though. We are seeing increasing numbers of investigator led, public funded, high quality trials and that has to be a good thing for clinicians and patients.
Why this matters
There are some cases that stick in the memory. Patient stories that don’t leave you and which make you reflect back on how the practice of medicine has changed, and is still changing. Many years ago Virchester I was a junior doc and med student at the time when we saw the start of ATLS style management in the UK. Virchester was at the forefront of developing systems and structures to manage our trauma patients.
I vividly remember a young man, a trauma patient who arrived in the ED talking. He was scared, he was in pain, but he was alive and the team that in a centre of excellence for trauma management there was lots that we could do. It was a blunt trauma mechanism with early signs of shock, but no head injury. We resuscitated him as best we could. Oxygen, bilateral chest drains and fluids…… lots of fluids. The initial 2 litres of 0.9% Saline were given without thought, all our major trauma patients got this. Our rationale of keeping the blood volume high and thus tissue perfusion high was our mantra. Those 2 litres were followed by a further litre of Gelofusine, and then when the blood pressure dropped again to 90mmHg, a further 1-2 iltres of Gelofusine as we fought to keep the MAP above 90mmHg. We had to do this as in or minds we had to maintain tissue perfusion and to stave off the concept of ‘irreversible shock’ whilst we waited for cross matched blood. Inevitably he got cold and acidotic and no doubt he had a coagulopathy, not that we could or would have been able to measure it. He crashed in resus , and despite the move to then using 0 negative blood he died. After arriving into the resus room with a GCS of 15 and an adequate cardiac output he died there in the resus room without ever making it to a CT scan or theatre. We drowned him in so much fluid that I remember taking blood samples towards the end that appeared almost straw like.
Some of the clinicians around at that time in emergency medicine, critical care, anaesthesia and surgery were truly groundbreaking in the way that changed our approach to critical illness and injury. They were and still are incredible clinicians whom I would trust the lives of my children to. They brought state of the art therapies such as Schoemaker optimisation regimes and Swan-Ganz catheters to the good folk of Virchester. These were good times when evidence based medicine was truly practiced. Or was it? Looking back it’s clear that much of what we did then appears dangerous now, and that may be the case today. Perhaps we will look back in 30 years and think ‘what the hell were we doing’, but that’s fine. Change in practice is good, dogmalysis is good, finding out that you are wrong and can do better is good. Arguably it is the most exciting thing about medicine. Uncertainty and change keep us agile and interested. Long may it continue.
Perhaps we should follow Simon Finfer’s lead and, to paraphrase Maya Angelou, do the best that we can until we know better, and then do better.
It's after midnight, I'm in the hospital, patient has undifferentiated shock. Thought processes and principles described in this paper are guiding investigation and management. https://t.co/pNMmNzVonK
— Simon Finfer (@icuresearch) August 19, 2018
- Subscribe to the blog (look top right for the link)
- Subscribe to our PODCAST on iTunes
- Follow us on twitter @stemlyns
- PLEASE Like us on Facebook
- Find out more about the St.Emlyn’s team
- Come join us at our conference in October 2018