Myth-busting: Lactated Ringers is safe in hyperkalemia, and is superior to NS. (2024)

_Introduction 0

_{Several months ago I gave a grand rounds on pH-guided resuscitation which was summarized in this post. This included a discussion that Lactated Ringers (LR) is safe in hyperkalemia. However, myth-busting is hard work. The dogma that LR should be avoided in hyperkalemia continues to replicate, both locally and on twitter. This myth drives me crazy because not only is it wrong, it's actually backwards. This post focuses on dispelling this misconception.}

_{The origins of the myth}0

_{LR has a potassium concentration of 4 mEq/L, whereas the potassium concentration of normal saline (NS) is zero. This has led to the common notion that LR should be avoided in a patient with hyperkalemia. I have been unable to find any evidence that LR actually causes or exacerbates hyperkalemia. Fear of LR in hyperkalemia appears to be entirely theoretical, due to the fact that LR contains potassium.}

_{Why the myth is wrong}0

This myth is wrong for three reasons. First of all, if a patient has hyperkalemia, then LR has a potassium concentration which is lowerthan the patient's potassium concentration. Administering LR to a patient with hyperkalemia will tend to pull the patient's potassium towards 4 mEq/L, and thereby decreasethe potassium level. This was explained well by Piper 2012:

_{thought often cited as a cause of hyperkalemia in those with renal dysfunction, infusion of LR with approximately 4 mEq/L potassium should not cause hyperkalemia. Even if the entirety of such a patient's plasma space was replaced with LR, the K+ concentration would not exceed the concentration of potassium in LR (4 mEq/L).}

_{Secondly, when one considers that the volume of distribution of potassium is greater than the extracellular fluid volume, it becomes clear that any infusion with a near-normal potassium concentration will have almost no effect on the serum potassium level. Consider, for example, a 70-kg man with a serum potassium concentration of 6 mEq/L and an extracellular fluid volume of 15 liters. Let's suppose that we infuse him with one liter of a solution containing 8 mEq/L potassium. His final serum potassium concentration will be a weighted average of 6 mEq/L multiplied by 15 liters and 8 mEq/L multiplied by 1 liter, which comes out to be 6.1 mEq/L. Thus, his potassium level only increases by 0.1 mEq/L, a barely measurable difference. Considering that potassium equilibrates between the intracellular and extracellular fluid spaces, its volume of distribution is much higher than the extracellular fluid volume and therefore the increase in potassium will be lower than 0.1 mEq/L (Huggins 1950; Winkler 1938).} Therefore, although a fluid withtwicethe potassium concentration of LR (8 mEq/L) could theoretically increase the serum potassium level, it would require a vast amount of such fluid to have any significant effect.0

_{Finally, the primary reason that this myth is wrong has to do with potassium shifting between the cells and the extracellular fluid. About 98% of the potassium in the body is present insidethe cells, with an intracellular potassium concentration of ~140 mEq/L. Therefore, even a tiny shift of potassium out of the cellular compartment will have a major effect on extracellular potassium levels. NS causes a non-anion gap metabolic acidosis, which shifts potassium out of cells, thereby increasing the potassium level. On the other hand, LR does not cause an acidosis, but instead may have a mild alkalinizing effect given that it contains the equivalent of 28 mEq/L of bicarbonate. Potassium shifts have a greater effect on the serum potassium than the actual concentration of potassium in the infused solution.}

_{Clinical evidence}

_{Three prospective double-blind randomized controlled trials have been performed comparing the effect of NS versus LR on potassium levels in patients with renal failure. These studies are a rare treat, because it's uncommon in critical care that we should have multiple confirmatory prospective RCTs.}

_{O'Malley 2005 performed a prospective, randomized, double-blind controlled trial of LR vs. NS for intraoperative IV fluid resuscitation during renal transplant surgery. The primary outcome of their study was postoperative creatinine concentration. The study was terminated early for safety reasons after interim analysis of data from 51 patients. Among the NS group 19% of patients developed potassium levels >6 mM intraoperatively compared to none in the LR group (p = 0.05; figure below). Patients in the NS group also had higher rates of metabolic acidosis requiring bicarbonate therapy (31% vs. 0%, p = 0.004). The authors concluded that giving large volumes of LR to patients undergoing renal transplant is safe and may be superior to NS.}

_{Khajavi 2008 replicated the O'Malley study. They performed a prospective, randomized, double-blind controlled trial of NS versus LR among 52 patients undergoing renal transplant surgery. The mean change in serum potassium during the procedure was +0.5 mEq/L in the NS group compared to -0.5 mEq/L in the LR group (p < 0.001; figure below). Patients in the NS group also had lower pH levels following surgery.}

Modi 2012 also replicated this study design with 74 patients undergoing renal transplantation. The average potassium among patients in the NS group increased by 0.37 mEq/L (p < 0.05), whereas there was no significant change in potassium among the LR group. At the end of the operation, potassium levels were higher in NS group:

_{There are some limitations to generalizing these results. All of these studies were done during surgical procedures, which could release potassium due to intra-operative tissue necrosis. Therefore, it remains unclear whether increases in potassium were due to surgery, normal saline infusion, or both factors combined. These studies involved administration of several liters of fluid, so administration of smaller volumes would likely to have less effect.}

_{What about Plasmalyte or Normosol?}

_{Plasmalyte and Normosol are balanced crystalloids of very similar composition to each other. Compared to LR, they have a slightly higher concentration of potassium (5 mEq/L vs. 4 mEq/L in LR) and are more alkalinizing (50 mEq/L bicarbonate equivalents vs. 28 mEq/L in LR). As discussed above, these fluids cannot cause hyperkalemia (specifically, they are not capable of causing a potassium level above 5 mEq/L). Furthermore, given the large volume of distribution of potassium, the exact potassium concentration of the infused fluid will have a negligible effect as long as it is near a physiologic range. The main determinant of changes in potassium concentration is likely changes in pH which lead to potassium shifts between the intracellular and extracellular spaces. Since these fluids are more alkalinizing than LR, when administered to a patient with metabolic acidosis they would be expected to cause a greater shift of potassium into cells, thereby reducing serum potassium to a greater extent. Bottom line? Although the potassium concentration of 5 mEq/L is cosmetically unappealing, these fluids should be safe in hyperkalemia and possibly superior to LR in patients with concomitant metabolic acidosis.}

_Conclusions

_{The myth that LR should be avoided in hyperkalemia is not only incorrect, it is probably backwards. For a hyperkalemic patient in renal failure, LR should be preferred over NS. LR has been proven to produce lower potassium levels en vivo. Understanding the effect that a crystalloid will have on serum potassium concentrations involves considering effects on acid-base physiology and intracellular potassium shifts, which are more important than the amount of potassium in the plastic bag.}

_{Image credits: Initial picture of potassium atom: http://commons.wikimedia.org/wiki/File:Electron_shell_019_Potassium_-_no_label.svg}