Fluid resuscitation for the cardiac surgical patient

Fluid resuscitation is a part of the "management encompassing resuscitation, initial and ongoing monitoring and supportive treatment" for "Cardiac surgery: including sternotomy for bypass grafting, valve surgery and aortic surgery", a Level 1 topic from Section 2.1.18 in the second edition of the CICM Syllabus for the Second Part Examination. This issue has never been challenged with a CICM exam question, perhaps because the examiners recognise that there is no standard against which they could test their candidates. was explored in Question 10 from the first paper of 2008, which had an extremely ordinary college answer associated with it, and which did not go into the management options.

That the patient will require some fluid following their procedure is a universally accepted fact, because:

• Surgery and cardiopulmonary bypass produce capillary leak, so there is a lot of fluid movement into the "third space"
• There are losses of blood, which will continue via the drains
• There is bypass-related cold diuresis
• Their stiff ventricles tend to function better with some extra preload
• The cold vasoconstricted patient will vasodilate and create more "unstressed volume" in their ciculatory system which is functionally equivalent to the loss of fluid

Which fluid? How much? This remains a subject of not so much debate as complacent indifference, as it appears we are all largely satisfied with leaving this to chance. Ramanan et al (2021) assessed Australian practice and determined that we mostly use balanced crystalloid, and Parke et al (2014) revealed that we use rather a lot of it: the usual Day 0 fluid balance was 2250 mL,  Clearly, knowing when to stop is a problem, and even though complex goal-directed algorithms do exist, the Parke study observed that 85% of the fluid boluses were initiated by the ICU nursing or junior medical staff, and arterial pressure was the usual trigger for this, even in patients who had complex haemodynamic monitoring attached. How these people then came to the conclusion that the patient is no longer in need of fluid is a mystery; presumably they were extubated and were rapidly weaned off vasopressors.

Assessment of fluid responsiveness in the cardiac surgical patient

The answer to the question "can I give this patient more fluid" is somewhat harder to answer in the post-op cardiac surgical patient, mostly because:

• The echo windows are terrible
• Nobody will let you lift their legs (because seriously)

However the other static and dynamic assessment measures remain valid. Or at least no less valid than they would usually be. Albano & Habana (2021) assessed some of these and determined that fluid-responders had some notable characteristics, as distinct from nonresponders:

• their pulse pressure variation was high (>12%)
• their CVP was low (<6)
• their PAWP was low (<12)

Yes, the wedge pressure. Apparently 68% of you still do use PA catheters. Even though nobody still uses the PAWP as a measure of preload anymore, the thermodilution cardiac output can be measured pre and post fluid bolus to determine whether there has been a benefit, and the user may interpret a modest change as a failure to respond. 

PAWP is likely an anachronism.  Back in the day, we would have used PAWP as a surrogate measure of LA pressure, which in turn is a surrogate measure of LV filling pressure, which in turn is an imperfect surrogate measure of LV filling volume. This string of inferences is fraught with error, and the decreasing use of PA catheters in the modern era has eroded the confidence of people how do not see them routinely, making them reluctant to handle these devices. Nevermind: if you do find yourself trapped in the land of PAWP-directed preload management, remember that the post-cardiotomy patient with their decreased ventricular compliance will need a higher filling pressure to achieve a given volume, and thus one may wish to run these people at a slightly higher PAWP. All the while remembering that PAWP is almost completely disconnected from preload.

The manufacturers of other advanced hemodynamic monitoring devices produce papers which would have us believe that their equipment can inform our use of fluids and thereby decrease the post-operative use of vasopressors and inotropes. Particularly, the GEDV is a calculated variable which has been widely trumpeted as a good number to base one's fluid management upon. One gives fluid challenges to increase the GEDV, and one stops either when they fail to increase the GEDV or the EVLWI increases, demonstrating that one has now flooded the lungs. Of course, as with all device-directed physiological manipulation, its a game of numbers and at no stage is anyone actually invited to look at the patient and assess their condition. Additionally, GEDV does not seem to stand up well in meta-analysis - its seems to be as bad as PAWP and CVP.