By Eric Walter, MD, MSc
Pulmonary and Critical Care Medicine, Northwest Permanente and Kaiser Sunnyside Medical Center, Portland, OR
Dr. Walter reports no financial relationships relevant to this field of study.
The use of intravenous (IV) fluids to replace lost volume is the sine qua non of resuscitation and one of the most common prescriptions in acute care. In 1832, Dr. Thomas Latta described the first use of IV fluids to treat a patient dying of cholera:
“Having inserted a tube into the basilic vein, cautiously — anxiously, I watched the effects; ounce after ounce was injected, but no visible change was produced. Still persevering, I thought she began to breathe less laboriously, soon the sharpened features, and sunken eye, and fallen jaw, pale and cold, bearing the manifest impress of death’s signet, began to glow with returning animation; the pulse, which had long ceased, returned to the wrist; at first small and quick, by degrees it became more and more distinct, fuller, slower, and firmer, and in the short space of half an hour, when six pints had been injected, she expressed in firm voice that she was free from all uneasiness, actually became jocular, and fancied all she needed was a little sleep; her extremities were warm, and every feature bore the aspect of comfort and health.”1
Today, the same life-saving effects of IV fluid administration are replicated over and over in intensive care units (ICUs) around the world. The use of IV fluids has become so universal that often little thought goes into the decision to “hang another bag.” However, since the day Dr. Latta first “dissolved from two to three drachms of muriate of soda and two scruples of the subcarbonate of soda in six pints of water”1 to make his solution, many different types of IV fluids have been developed. Unfortunately, there is little evidence to help ICU clinicians choose one type of fluid over another. The decision of which fluid to use is highly variable and predominantly driven by local practice rather than patient or fluid characteristics.2
There are two general classes of IV fluids: crystalloids and colloids. Crystalloids contain solutions of ions that easily pass through a healthy semi-permeable membrane, while colloids contain suspensions of large molecular weight molecules that poorly diffuse through a healthy semi-permeable membrane. For decades, the relative merits of crystalloids and colloids have been debated, but multiple trials and systematic reviews have concluded that neither fluid is safer nor more effective than the other for the resuscitation of critically ill patients.3 In the United States, crystalloids are used far more often than colloids in the ICU.2 There are two primary categories of crystalloids: normal saline (NS) and balanced electrolyte solutions. NS consists of 0.9% saline, while balanced electrolyte solutions contain a buffer. However, both of these names are misleading.
NORMAL SALINE
NS is the most commonly used crystalloid worldwide. In 2011, more than 200 million liters were sold in the United States alone.4 It was developed in the 1880s based on in vitro studies suggesting the salt concentration of blood was 0.9%. This was incorrect. The true salt concentration of human blood is actually 0.6%, with a normal plasma sodium concentration of 135-144 mmol/L. Thus, “normal saline,” with a sodium concentration of 154 mmol/L, is actually hypertonic.3,5-6 NS is also acidotic with a pH that ranges from 4.5 to 7.0 (See Table 1). This appears to be related to the polyvinyl chloride bags used for packaging, as the pH is 7.0 when stored in a glass bottle.5
BALANCED ELECTROLYTE SOLUTIONS
In the 1920s, Alexis Hartmann treated children with gastroenteritis with an alkalinized salt solution that had originally been used in Sydney Ringer’s cardiac physiology lab.6 Hartmann’s solution, Ringers lactate, and Ringers acetate are modifications of this fluid and some of the most commonly used balanced electrolyte solutions used today. “Balanced solutions” are described as having ionic compositions similar to plasma. However, none of these solutions are truly balanced.3,5-6 Most are relatively hypotonic due to sodium concentrations less than extracellular fluid, and they all contain buffers that differ from the bicarbonate and chloride found in plasma (See Table 1). The most common buffers are lactate, acetate, or gluconate.
COMPARING NORMAL SALINE AND BALANCED ELECTROLYTE SOLUTIONS
Crystalloids are inexpensive, stable, and widely available, but they do have shortcomings. Resuscitation with large volumes of NS frequently causes a hyperchloremic metabolic acidosis. Clinical studies in animals and human volunteers have shown that NS infusions may cause decreased renal blood flow, delayed micturition, acute kidney injury (AKI), and may affect immune function.3,5-7 It is not known if this is due to hyperchloremia, the strong ion difference (concentration difference between fully dissociated cations and fully dissociated anions), or some other factor.8 On the other hand, large volumes of balanced solutions may result in hyperlactatemia, metabolic alkalosis, hypotonicity, and cardiotoxicity.3 Some balanced solutions cannot be infused in the same IV line as blood products because they contain calcium, and there is a concern for microthrombi generation,3 although others have argued against this.9 Until recently, there has been very little data to say whether one solution is better than the other.
Much of the data comparing NS and balanced solutions has been observational, and some of the larger studies are reviewed here. In a single-institution, prospective, before-and-after study, Yunos et al compared administration of a “chloride-liberal” (control period) IV fluid strategy to a “chloride-restrictive” (intervention period) strategy.10 During the 6-month chloride-liberal period, NS was the primary fluid used in the ICU. This was followed by a 6-month phase-out period for staff education and to ensure that the intervention period would occur during the same season as the control period. The authors then preferentially used balanced solutions for 6 months. During this chloride-restrictive period, NS could only be used for specific conditions (e.g., hyponatremia, traumatic brain injury, and cerebral edema). There were no significant differences in baseline characteristics between patients treated during the control (n = 760) and intervention (n = 773) periods. Implementing the chloride-restrictive strategy decreased NS use from 2411 liters to 52 liters and increased Hartmann solution use from 469 liters to 3205 liters. The incidence of AKI (odds ratio [OR], 0.52; 95% confidence interval [CI], 0.37-0.75) and use of renal replacement therapy (OR, 0.52; 95% CI, 0.33-0.81) both decreased by nearly 50%. There were no significant differences in length of stay or mortality.
In a retrospective study, Shaw et al compared NS to Plasma-Lyte (a proprietary balanced solution) using a national hospital claims database to evaluate a composite outcome of postoperative complications following major abdominal surgery.4 Patients who received NS exclusively (n = 30,994) on the day of surgery were compared to those who exclusively received Plasma-Lyte A or Plasma-Lyte 148 (n = 926) on the day of surgery. At baseline, the NS group appeared sicker. They had more comorbidities, such as heart failure, diabetes, and renal failure, and were more often admitted through the emergency department. To attempt to control for these baseline differences, propensity matching was performed. The use of NS was associated with one or more major complication compared to the use of Plasma-Lyte (33.7% vs 23%, P < 0.001). This association remained significant after a propensity-matched adjustment (OR, 0.79; 95% CI, 0.66-0.97). Renal replacement therapy was more common in the NS group (1.0% vs 4.8%, P < 0.001). These results suggest a benefit to Plasma-Lyte, but given the baseline differences between groups, residual confounding remains a significant concern with this study. The manufacturer of Plasma-Lyte funded the study.
Using the same national hospital claims database, Raghunathan et al compared NS and balanced fluids for resuscitation of septic patients in two studies. Patients treated with NS were compared to patients treated with balanced fluids (predominantly Ringer’s lactate).11,12 Similar to the study by Shaw,4 there were significant differences between groups at baseline. The NS groups were more likely to have heart failure, chronic renal failure, hypertension, and diabetes. They also received less fluids overall. Propensity matching and other methods of risk adjustment were used to attempt to adjust for the imbalances between groups. Both studies reported an approximately 10-15% relative increase in mortality with the use of NS compared to balanced fluids. However, as before, residual confounding remains a major question. Furthermore, no increase in any secondary outcomes, including acute renal failure, occurred, so the mechanism leading to increased mortality is not clear.
A meta-analysis comparing high-chloride fluids (NS in all studies) with low-chloride fluids (Ringer’s lactate, Hartmann’s solution, or Plasma-Lyte) has been published.13 The analysis combined results from observational studies and several controlled clinical trials. The conclusion was that resuscitation with NS was associated with a significantly higher risk of AKI (relative risk [RR], 1.64; 95% CI, 1.27-2.13) without a difference in mortality (RR, 1.13; 95% CI, 0.92-1.39). These results were highly driven by the large before-and-after study by Yunos discussed previously (77% of the weight).10 When this study was excluded, the increased risk of AKI was still present, although no longer statistically significant.
Since publication of this meta-analysis, results from a large randomized clinical trial have become available.14 The SPLIT trial was a double-blind, cluster randomized trial where individual ICUs were randomized to either the sole use of NS (n = 1067 patients) or Plasma-Lyte 148 (n = 1025 patients) in a double-crossover protocol. At baseline, both the NS and Plasma-Lyte groups were well-matched. Of note, patients were quite healthy (~85% without any comorbidities) and most patients were admitted to the ICU following elective surgery (most often cardiovascular surgery). The median volume of fluids administered was modest at only 2000 mL per patient in both groups. No significant differences were found in the development of AKI (9.2% in the saline group vs 9.6% in the Plasma-Lyte group, P = 0.77), need for renal replacement therapy (3.4% vs 3.3%, P = 0.91) or in-hospital mortality (8.6% vs 7.6%, P = 0.40). These results contrast with many prior studies. A criticism of this trial is that the small volume of fluid administered may not have been enough to show any effect. Furthermore, this healthy population of mostly elective surgery patients may have been at very low risk of complications. On the other hand, it may be true that one crystalloid is not any better than the other. This would not be the first time in critical care fluid trials where a randomized controlled trial refuted prior observational studies and small trials.15 However, given the valid criticisms of SPLIT, there are ongoing calls for more randomized, controlled trials.
SUMMARY
IV fluids have come a long way in a short time since Latta’s first description. Clearly, they have proven to be a life-saving intervention for many patients. However, the use of these fluids should not be taken lightly and deserves the same critical evaluation afforded other “drugs.” There is emerging evidence suggesting that balanced salt solutions are safer than NS. Definitive evidence is still lacking, but it may be time to reconsider the role of NS. Given the ubiquitous use of IV fluids worldwide, this question deserves to be answered definitively with large randomized, controlled trials. For even if the risk or benefit differences are small, the public health impact could be large.
REFERENCES
-
Latta T. Documents communicated by the central board of health, London, relative to the treatment of cholera by the copious injection of aqueous and saline fluid into the veins. Lancet 1832;18:274-280.
-
Finfer S, et al. Resuscitation fluid use in critically ill adults: An international cross-sectional study in 391 intensive care units. Crit Care 2010;14:R185.
-
Myburgh JA, Mythen MG. Resuscitation fluids. N Engl J Med 2013;369:1243-1251.
-
Shaw AD, et al. Major complications, mortality, and resource utilization after open abdominal surgery: 0.9% saline compared to Plasma-Lyte. Ann Surg 2012;255:821-829.
-
Moritz ML, Ayus JC. Maintenance intravenous fluids in acutely ill patients. N Engl J Med 2015;373:1350-1360.
-
Myburgh JA. Fluid resuscitation in acute medicine: What is the current situation? J Intern Med 2015;277:58-68.
-
Marik PE. Iatrogenic salt water drowning and the hazards of a high central venous pressure. Ann Intensive Care 2014;4:21.
-
Elbers P, et al. Chloride-restrictive fluid administration and incidence of acute kidney injury. JAMA 2013;309:542-543.
-
Lorenzo M, et al. Can Ringer’s lactate be used safely with blood transfusions? Am J Surg 1998;175:308-310.
-
Yunos NM, et al. Association between a chloride-liberal vs chloride-restrictive intravenous fluid administration strategy and kidney injury in critically ill adults. JAMA 2012;308:1566-1572.
-
Raghunathan K, et al. Association between the choice of IV crystalloid and in-hospital mortality among critically ill adults with sepsis*. Crit Care Med 2014;42:1585-1591.
-
Raghunathan K, et al. Association between initial fluid choice and subsequent in-hospital mortality during the resuscitation of adults with septic shock. Anesthesiology 2015;123:1385-1393.
-
Krajewski ML, et al. Meta-analysis of high- versus low-chloride content in perioperative and critical care fluid resuscitation. Br J Surg 2015;102:24-36.
-
Young P, et al. Effect of a buffered crystalloid solution vs saline on acute kidney injury among patients in the intensive care unit: The SPLIT Randomized Clinical Trial. JAMA 2015;314:1701-1710.
-
Myburgh JA, et al. Hydroxyethyl starch or saline for fluid resuscitation in intensive care. N Engl J Med 2012;367:1901-1911.
