The Challenge of Defining the "Classic Fluid Challenge"

Volume 28 | Issue 3 | Fall 2017 

Messina A, Longhini F, Coppo C, Pagni A, Lungu R, Ronco C, Cattaneo MA, Dore S, Sotgiu G, and Navalesi P. Use of the fluid challenge in critically ill adult patients: a systematic review. Anesthesia and Analgesia 2017; 125: 1532-43.

As much as we want to believe that most medical interventions are carefully researched and based on the best possible evidence, a large proportion of what we do for patients is based on tradition, consensus, or what some refer to as “eminence-based medicine,” i.e. emulating what the most respected physicians do. The fluid challenge (FC) is a diagnostic and therapeutic intervention that has been used for years in critical care as an easy way to determine fluid responsiveness during resuscitation. In general terms, an FC attempts to assess whether a given patient’s hemodynamics will improve with further fluid administration.[i] Many clinicians refer to the use of a “classic” or “typical” fluid challenge, but it seems that this paradigmatic FC has never really been conclusively defined. The FENICE Trial, published in 2015, attempted to survey practices involving the FC and found that there was a great deal of variability regarding the administration of FC and how the results were evaluated.[ii] Variation in the way an FC is performed can affect its sensitivity and specificity and lead to different therapeutic choices. While fluid administration is essential to maintaining organ perfusion, volume overload can also be detrimental, so inconsistencies in assessment and decision-making can have a significant effect on outcomes.

As we know too well, it is common that real-world practice strays from what is recommended in the medical literature. So is the variability in FC administration a result of translation from evidence to practice, or is it present in the literature as well? Messina and colleagues tried to address this question in the article under discussion, published in November 2017. The researchers reviewed studies published on the FC within the past 20 years and assessed them with regard to the amount and type of fluid administered, duration of infusion, hemodynamic variables measured and determination of fluid responsiveness, and safety limits. They defined FC as the infusion of a specific amount of fluid in a fixed time and searched MEDLINE, EMBASE, and the Cochrane Database of Systematic Reviews to find appropriate studies published between January 1994 and December 2014. Patients were divided into subgroups based on underlying disease process. The researchers surveyed the practices of the studies and performed a logistic regression to assess the relationship between the percentage of FC “responders” in each study and several independent covariates.

A total of 71 studies were identified that included a total of 3617 patients. Most subjects were receiving mechanical ventilation, although 8.5% of the studies examined enrolled only spontaneously breathing patients. The most common indicators for an FC were hypotension (67.6% of studies), oliguria (52.1%), and physician judgment (49.3%). Only 5.6% of the studies had a predetermined safety limit for stopping an FC.

The median of the mean fluid volume administered in the 17% of studies that reported weight-based dosage was 7 mL/kg; of the remaining studies, the majority (77.5%) infused 500 mL of fluid. Colloids were used in 62% of the studies with the majority utilizing 6% hydroxyethyl starch (HES), and 37.5% administered crystalloids. When studies were grouped by years published, the proportion using colloids decreased over time. There was substantial variation in the duration of fluid administration, with 30 minutes being the most popular choice (45.1%), but 10, 15, and 20 minutes also being prevalent.

The mean of the mean rates of FC “responders” found across the studies was 52%. An increase in cardiac index or output was used to assess fluid responsiveness in 62% of studies, most commonly using a 15% increase as the criterion. 31% of studies used stroke volume or stroke volume index, again with a 15% increase being the most common standard used to assess response. Not surprisingly, in the 45 studies that reported arterial pressure variation, the amount of variation prior to FC was higher in responders than nonresponders (11.5 ± 5.4% vs. 6.1 ± 3.9%, p < .001).

When looking at subgroups, the only two subgroups including studies that enrolled more than 75% of patients with the same diagnosis were those examining patients with sepsis and postsurgical patients, so these were the only two included in the regression analysis. The researchers found that none of the variables examined (sepsis, postoperative state, hypotension, oliguria, colloid use, or time of FC administration) were significantly correlated with the number of responders in the study. Notably, there was a trend towards studies with shorter time of administration having a greater proportion of responders, with p < 0.7.

So, what are we to make of this study? Does it add anything new to the not-very-controversial observation that there is variation in exactly what a fluid challenge means to different practitioners? One observation is that in the studies reviewed, only about half of the patients responded to an FC, implying that our understanding of who will respond is incomplete, and some of the triggers used (i.e. hypotension) may not be very specific for fluid-responsiveness. It would also be helpful to have some more detailed dose-response studies to better examine different bolus types and amounts within a given patient population as opposed to across different studies. It is interesting that there was a trend towards having a greater number of responders in studies that gave faster fluid challenges; some studies have shown that the effect of a rapid crystalloid bolus dissipates within 10 minutes of administration.[iii] This again reinforces the need for more within-study evaluations of technique.

Another issue centers on the changes in FC practice over time. The majority of papers in the review utilized colloids, particularly HESs; these have been less in favor since the SAFE study, which implied that colloids were no more effective than crystalloids in resuscitation and could be harmful in certain populations.[iv] Other studies have suggested a particularly greater risk of renal dysfunction and death with HESs compared to other fluids, and the most recent Surviving Sepsis Campaign guidelines recommend against their use.[v] In the 2015 FENICE survey, by contrast, crystalloids were preferred in 74% of cases.² This evolution in practice may bring into question whether the results from earlier studies are applicable to modern critical care. Clinical practice and scientific evidence both suffer from a lag between data’s creation and its implementation, and the two strains may not progress at the same rate.

While medical decision-making always involves an element of subjectivity, this should not dissuade us from pushing forward and trying to better delineate best practices and standards of care. On the other hand, there are limits on how many resources we can devote to a given clinical problem. Not every corner of the biomedical world can be examined in minute detail. We may still be frustrated by the lack of standardization of the fluid challenge, but 500 mL “pretty fast” may have to suffice for now.