What Minimal Volume of Intravenous Fluid Challenges the Heart?
Jon-Emile S. Kenny MD [@heart_lung]
“I find real life challenging enough. If I get to the dry cleaners without having a huge fight, that’s challenging enough.”
-Fran Lebowitz
Background
Quite recently, the BaSICS investigators evaluated the infusion rate of a fluid bolus on patient outcome. As considered, however, contrast should be made between a fluid bolus, which is therapeutic and a challenge, which is diagnostic. For what are intravenous fluids diagnostic? As described in innumerable reviews, a fluid challenge may be executed to test cardiac fluid tolerance. For example, should a fluid challenge fail to raise stroke volume [SV], a state of physiological preload intolerance exists and additional volume is arguably harmful; further attempts to increase tissue perfusion might then be best achieved with vasoactives.
Yet both the dose [i.e., volume] and timing [i.e., rate] of a fluid challenge is poorly defined. Indeed, a recent consensus statement suggested a fluid challenge be 500 mL infused in less than 30 minutes. Yet, studies of fluid challenges have actually investigated greater rates of infusion including 50 mL in 10 seconds! As well, the volume of administration might be much lower than 500 mL, with previous research indicating that 4 mL/kg of crystalloid was an optimal volume needed to increase mean systemic filling pressure.
With this, Barthelemy and colleagues have recently published a clever study in critically-ill patients with circulatory failure; the ‘dose-response’ curve of intravenous crystalloid is described as well as the optimal volume needed to test cardiac preload reserve.
What they did
Adult, critically-ill patients monitored with esophageal Doppler were enrolled when the treating physician performed a 500 mL crystalloid challenge for acute circulatory failure. Patients were excluded if they were not an adult, if they were pregnant, had dysrhythmia, or had poor Doppler signal quality.
50 mL of crystalloid was infused over 30 seconds, repeated 10 times, every 1.5 minutes up to a total of 500 mL. SV was measured at baseline and 1 minute after each aliquot of 50 mL.
The primary outcome was identification of the optimal volume for assessing fluid responsiveness. Fluid responsiveness was defined as an increase in SV of at least 15% from baseline to the end of the 500 mL protocol. The secondary outcomes were the shape of the ‘dose-response’ curve for successive fluid challenges and the identification of the fluid volumes associated with the probability of positive response in 50% [i.e., ED50] and in 90% [i.e., ED90] of the patients who were ‘responders.’
What they found
45 critically-ill patients were included with a 28-day mortality rate of 29%; distributive shock was the most common cause of circulatory failure with sepsis being the most common cause of distributive physiology. All patients were invasively-ventilated at the time of study and 69% were receiving norepinephrine. Following the 500 mL protocol, 80% of the patients increased SV by at least 15%.
The area under the receiver operator curve for each aliquot of 50 mL increased up to a total of 250 mL administered with a value of 0.93 for predicting fluid responsiveness; no statistically-significant increase in the AUROC occurred beyond 250 mL. The median ED50 was 156 mL and the median ED90 was 312 mL. These values corresponded with weight-adjusted volumes of 3 mL/kg, 2.0 mL/kg and 4.2 mL/kg, respectively.
Thoughts
Barthelemy and colleagues report an inventive and necessary study for those interested in resuscitation physiology. While a single centre study of only 45 patients, the results nevertheless provide much-needed clarity for future consensus statements and management guidelines. The strength of their findings is its consistency with previous work using separate methodology. The volume of 250 mL lines well with the 4 mL/kg noted to increase mean systemic filling pressure in stable cardiothoracic surgical patients. Recall that in Guyton’s model, an increase in mean systemic pressure enhances venous return, which interacts with the heart by stretching cardiac myocytes.
One aspect of their results that surprised me was the fraction of patients with preload reserve. They observed that 80% of their participants had a physiologically-significant increase in SV at the end of the 500 mL protocol. This is much larger than the anticipated 50/50 split between non-responders and responders that is usually observed in the critically-ill. This is especially salient because the esophageal Doppler that they employed assumes a fixed diameter of the aorta. In hypotensive patients, this kind of monitor can miss flow augmentation if some of the volume-per-beat is expressed in the diameter domain in addition to the velocity domain; the sensitivity of esophageal Doppler is enhanced [i.e., false negatives reduced] when diameter is measured in the critically-ill. This implies that some of their ‘non-responders’ may have been misclassified and that there was actually more than 80% with preload reserve.
As well, questions remain about how to specifically operationalize this mini fluid challenge at the bedside – namely the rate of infusion. The authors designed the entire protocol to mimic 500 mL given over 15 minutes, but there is another way to look at the infusion rate of their investigation. The 50 mL aliquot was physically infused over 30 seconds; that is, crystalloid was actually delivered at a rate of 100 mL per minute. This rate is effectively less than the initial study of the mini-fluid challenge by Muller and colleagues who infused 100 mL of colloid over 1 minute and of Wu et al. who studied 50 mL of crystalloid in 10 seconds.
Might the infusion rate affect the cardiac myocyte response to volume? We know that purely viscous structures increase their stress [e.g., trans-mural pressure] in direct proportion to strain velocity, while purely elastic structures do not. As in the lung, cardiac myocytes are viscoelastic, such that strain velocity [i.e., the rate at which volume changes] probably does have some effect on the trans-mural wall stress. In theory, it is the wall stress that triggers stretch-induced calcium sensitivity [i.e., the Frank-Starling mechanism]. Therefore, the rate at which volume is added to the circulatory system could very well mediate the aforementioned; only further investigation will edify.
Best,
JE
Dr. Kenny is the cofounder and Chief Medical Officer of Flosonics Medical; he is also the creator and author of a free hemodynamic curriculum at heart-lung.org. Download his free textbook here.