For a ubiquitous intervention such as intravenous fluid resuscitation, it’s a surprise to see reassessment continuing apace. Since the inception of early goal-directed therapy, emergency departments (EDs) have been dumping fluids, typically 0.9 percent sodium chloride solution, into patients in ever-increasing quantities. Even as early goal-directed therapy has fallen out of favor, its ongoing evolution has discarded only the most invasive bits, while keeping early antibiotics and aggressive resuscitation.

0.9 percent sodium chloride is hardly physiologically equivalent to the intravascular circulation, nor does it specifically address electrolyte losses or variance in the setting of critical illness. The popularized treatment alternative remains so-called balanced crystalloid solutions. These solutions typically include some small amount of sodium chloride, but may also include sodium gluconate, sodium acetate, sodium lactate, calcium chloride, potassium chloride, or potentially even magnesium chloride. The pH of these solutions is closer to the normal physiologic value of 7.4, rather than the pH of roughly 5.5 seen with 0.9 percent sodium chloride.

Many trials have been published comparing the two types of electrolyte solutions, including those in the ED and in the intensive care unit (ICU). Just this spring, such a comparison has been published, and again, no statistically significant difference could be detected between 0.9 percent saline and a balanced solution, in this case Plasma-Lyte 148.¹ The most notable difference in thinking following this most recent randomized controlled trial is its accompanying updated systematic review and meta-analysis of the same topic.² Taking stock of the entirety of all the high-quality randomized controlled trials, the authors of the meta-analysis pooled nearly 35,000 ICU patients together to develop a more precise estimate of any treatment effect associated with use of balanced crystalloid solutions. Despite this larger sample size, no statistically significant difference in mortality could be identified.

However, the story does not quite end there. The risk ratio from their pooled low-bias study cohort was 0.96 [95 percent CI 0.91 to 1.01], non-significantly favoring the balanced solutions. The authors then utilize a Bayesian approach to assessing the probability that balanced fluid solutions are superior. Based on a vague assumption that the prevailing observational evidence and opinion is insufficient to strongly tilt the results of any analysis, the authors conclude there is an 89.5 percent posterior probability that balanced crystalloids reduce mortality.

Considering that this analysis encompasses 35,000 patients and is only able to eke out such a minor advantage to treatment with balanced fluids, the absolute magnitude of any treatment effect must be quite small. Then, these pooled studies are all from the ICU, limiting their generalizability to the ED. The simple way out is to say, “It doesn’t matter, and, if it matters, it only matters a tiny amount.” Conversely, sepsis is implicated as a leading cause of death worldwide, and significant volumes of intravenous fluid are given in emergency departments each day. Even a miniscule relative improvement in mortality—a rather important patient-oriented outcome—becomes a significant absolute excess in survival when multiplied across millions of lives. Whenever the option exists to choose a balanced solution, it is likely the ever-so-slightly better choice.

How Much Fluid to Administer

The next question, however, having made the choice of fluid, remains precisely how much to give. This is the question addressed by the Conservative Versus Liberal Approach to Fluid Therapy of Septic Shock in Intensive Care (CLASSIC) trial.³ In this trial, patients admitted to the ICU were randomized to standard care or a conservative strategy in which 250- or 500-mL fluid boluses were permitted only for objective evidence of ongoing and worsening severe hypoperfusion. The difference in fluid administered in the ICU despite these restrictions, however, was only about 1.5 liters over the first five days. Ultimately, the authors could not identify any specific advantage of one treatment strategy versus another.

Tucked within these data, however, were some interesting secondary analyses with potential relevance to the emergency physician. On average, patients had already received three liters of fluid prior to enrollment in the trial. However, approximately one-third of their cohort had received less than 30 mL/kg of fluid upon enrollment, while the remainder had received greater than that amount. In patients having received the lowest fluid resuscitation per body weight there was a 5.3 percent mortality advantage to allowing the standard liberal strategy in the ICU, while those aggressively resuscitated prior to enrollment displayed a 2.1 percent advantage to a restrictive strategy. Neither of these advantages were statistically significant, and might have been affected by multiple potential confounding biases, but it does fit with a reassuring narrative: the fluid resuscitation commonplace in the emergency department may indeed be the necessary first step. Further data relevant specifically to early fluid administration is likely to be available from the Crystalloid Liberal or Vasopressors Early Resuscitation in Sepsis (CLOVERS) trial, which has seen early termination of enrollment in May 2022.

It isn‘t all bad news for 0.9 percent saline, however, as the evidence works in its favor in the Resuscitation with Blood Products in Patients with Trauma-Related Haemorrhagic Shock Receiving Prehospital Care (RePHILL) trial.⁴ This is a randomized controlled trial in which trauma patients with hypotension thought to be related to traumatic hemorrhage received either prehospital blood and plasma or 0.9 percent sodium chloride. The total volume of fluids to be administered in the trial protocol was either two units of packed red blood cells combined with two units of lyophilized plasma or one liter of sodium chloride.

In this physician-led prehospital service in the United Kingdom, the authors were able to enroll 432 patients prior to the onset of the SARS-CoV-2 (COVID-19) global pandemic. The enrolled population was primarily young and male, and the majority of patients were included as a result of severe injuries sustained in motor vehicle collisions. The primary outcome was a composite of mortality and inability to clear lactate within two hours of randomization, but the more likely relevant patient-oriented outcome is mortality alone. In this quite severely injured cohort with prehospital hypotension, overall mortality was 43 percent in those receiving blood products and 45 percent in those receiving saline.

As with any snapshot of evidence, this remains only a piece of the puzzle with respect to the utility of prehospital blood products in trauma. These authors do not parse out specific causes of death, and generally most patients were not yet in hemorrhagic shock on arrival at the hospital. This is presumably linked to the short transport times involved from the scene to definitive care, reducing the likelihood of exsanguination prior to hospital arrival.

The RePHILL trial fits in with other trials of prehospital plasma or prehospital blood products. It informs resource management for urban prehospital systems with short transport times. Replacing blood with blood has obvious face validity, but many other clinical scenarios have found conservative transfusion thresholds are not harmful. Early in the course of traumatic hemorrhage, with prehospital shock but substantial remaining reserve, aggressive early use of blood products may not be providing the hoped-for survival advantage. Future trials ought to illuminate at which point the resuscitation can switch from crystalloid to blood products.

Dr. Radecki is an emergency physician and informatician with Christchurch Hospital in Christchurch, New Zealand. He is the Annals of Emergency Medicine podcast co-host and Journal Club editor and can be found on Twitter @emlitofnote.

References

1. Brown J, Lane A, Cooper C, Vassar M. The results of randomized controlled trials in emergency medicine are frequently fragile. Ann Emerg Med. 2019;73(6):565-576.
2. Parish AJ, Yuan DMK, Raggi JR, Omotoso OO, West JR, Ioannidis JPA. An umbrella review of effect size, bias, and power across meta‐analyses in emergency medicine. Academic Emergency Medicine. 2021;28(12):1379-1388.
3. Fanaroff AC, Califf RM, Windecker S, Smith SC, Lopes RD. Levels of evidence supporting american college of cardiology/american heart association and european society of cardiology guidelines, 2008-2018. JAMA. 2019;321(11):1069.
4. Ibrahim SA, Reynolds KA, Poon E, Alam M. The evidence base for US joint commission hospital accreditation standards: cross sectional study. BMJ. Published online June 23, 2022:e063064. Hill, K. How Target figured out a teen girl was pregnant before her father did. Forbes. https://www.forbes.com/sites/kashmirhill/2012/02/16/how-target-figured-out-a-teen-girl-was-pregnant-before-her-father-did. Published February 16, 2012. Accessed September 22, 2022.