In order to choose the best fluid therapy for DKA, a thorough understanding of the electrolyte contents of crystalloid solutions is essential.

The composition of Plasmalyte is similar to that of plasma, with a pH range of 6.0 to 7.4 and an osmolarity of 294 mOsmol/L. It contains sodium, potassium, chloride, magnesium, acetate, and gluconate. The concentrations are as follows: potassium, 5 mmol/L; sodium, 140 mmol/L; chloride, 98 mmol/L; magnesium, 1.5 mmol/L; acetate, 27 mmol/L; and gluconate, 23 mmol/L [11,16].

Normal saline is an isotonic crystalloid solution with an osmolarity of 308 mOsmol/L and a pH range of 4.5 to 7.0. Furthermore, it has an equal amount of both sodium and chloride ions (154 mmol/L). Potassium, magnesium, and bicarbonate are absent, which can have an impact on electrolyte imbalances and the acid-base state in DKA [11,17].

Ringer’s lactate solution is an isotonic balanced crystalloid solution containing various essential electrolytes like sodium, chloride, potassium, calcium, and lactate. It has an osmolarity of 273 mOsm/L and a slightly acidic pH of approximately 6.5. The solution consists of the following specific amounts of these electrolytes: sodium, 130.5 mmol/L; potassium, 4.02 mmol/L; calcium, 0.67 mmol/L; chloride, 109.6 mmol/L; and lactate, 28 mmol/L [18,19].

In a randomized controlled trial, Ringer’s lactate and 0.9% sodium chloride were compared in the management of DKA [20]. The group receiving 0.9% sodium chloride took 683 minutes to reach a pH of 7.32, while the Ringer’s lactate group took 540 minutes. The duration for Ringer’s lactate to reach a blood glucose level of 14 mmol/L was notably longer than that of 0.9% sodium chloride. Both groups had comparable median hospital stays of 7 days and there were no fatalities. According to the study, Ringer’s lactate did not outperform 0.9% sodium chloride in resolving acidosis in DKA patients and even led to delayed glycemic recovery.

Plasmalyte is a well-balanced intravenous fluid comprising sodium, potassium, magnesium, calcium ions, and anions like acetate. Large administration of chloride-containing fluids such as normal saline leads to hyperchloremic metabolic acidosis. The ability of a balanced electrolyte solution to prevent hyperchloremic metabolic acidosis can be understood by examining the differences in pH and chloride content between serum, normal saline and balanced electrolyte solution. Normal saline has a pH of around 5.5 and a chloride content of 154 mmol/L. In contrast, a balanced electrolyte solution (specifically plasma-lyte A) has a pH of 7.4 and chloride content of 98 mmol/L, which is more similar to that of human plasma (94–111 mmol/L) [12,21]. The electrical neutrality of the serum is maintained by balancing positive and negative ions. When normal saline is used in larger quantities, serum chloride increases, which leads to the loss of an equivalent amount of bicarbonate ions to maintain electrical neutrality. This change causes hyperchloremic metabolic acidosis. Plasmalyte, in contrast, does not cause a decrease in bicarbonate levels [12].

A study conducted by Mahler et al. [12] investigated the effects of resuscitating DKA patients with a balanced electrolyte solution compared to those with normal saline. They found that patients who received balanced electrolyte solution had lower serum chloride levels and high serum bicarbonate levels after resuscitation; these results suggest a potential preventive effect of balanced electrolyte solution against hyperchloremic metabolic acidosis. In contrast, the normal saline group exhibited higher post-resuscitation serum chloride levels, aligning with existing literature that suggests normal saline administration can contribute to hyperchloremic metabolic acidosis in patients with DKA.

The study conducted by Mahler et al. [12] reinforced the idea that the use of balanced crystalloids is effective in preventing hyperchloremic metabolic acidosis, showcasing its strength. However, this study was limited by its small sample size, single-center design and lower baseline serum chloride levels in the sodium chloride group compared to those in the balanced crystalloids group.

Emerging evidence from recent preclinical and early clinical investigations has raised concerns about the potential adverse effects associated with the administration of normal saline [13,22]. Several reports have hypothesized that there is a connection between hyperchloremia and acute kidney injury [22]. Studies conducted on critically ill adults have consistently shown an increased risk of acute kidney injury and a greater need for renal replacement therapy when chloride liberal fluids are utilized [14,22]. In contrast, studies have consistently indicated a lower prevalence of hyperchloremic acidosis and renal injury, leading to improved patient outcomes, when balanced chloride-restrictive fluids are used. While the precise clinical implications of normal saline infusion are not yet fully understood, there is growing evidence indicating that its use may heighten the risk of kidney injury and hinder recovery from severe illness, potentially due to the induction of metabolic acidosis [13]. However, there is still no definitive relationship established between hyperchloremia and acute kidney injury.

The existing literature indicates that hyperchloremic metabolic acidosis can cause disruptions that potentially have negative effects on patient outcomes. Animal studies on sepsis have shown less metabolic acidosis, reduced inflammatory cytokines and improved survival rates with balanced electrolyte solution compared to those with normal saline. Hyperchloremic metabolic acidosis has also been associated with increased blood product transfusion in postoperative patients and impaired renal function [12].

While Plasmalyte has demonstrated effectiveness in treating DKA, the assessment of its safety in patients with severe DKA has not been conducted formally [14]. Plasmalyte contains acetate with a concentration of 27 mmol/L. Acetate acts as a precursor for acetoacetate, which is one of the ketone bodies involved in DKA. In canines, acetate infusions have been shown to raise acetoacetate levels. Rat studies have demonstrated that liver mitochondria can convert acetate into acetoacetate. In humans, there is some evidence suggesting that acetate contributes to ketogenesis, particularly in the context of hemodialysis. Nevertheless, it remains unknown whether a similar metabolically significant process occurs in patients with DKA and warrants further investigation [11].

Nonetheless, it is important to acknowledge that balanced crystalloids carry theoretical risks in DKA treatment, including the potential for alkalosis and hyperkalemia [13,21]. Additionally, the comparative effects of balanced crystalloids and saline in this context are not well understood.

Homogenous endpoints were found in two studies each: median hospital length stay in studies by Ramamnan et al. [11] and Attokaran et al. [14] (Table 2) and time to DKA resolution in studies by Self et al. [13] and Tsui et al. [15] (Table 3).

According to the SCOPE-DKA (Sodium Chloride or Plasmalyte-148 Evaluation in Severe Diabetic Ketoacidosis) trial by Ramanan et al. [11], although the length of stay was shorter in patients receiving Plasmalyte than it was in those receiving normal saline, the rate of readmission was higher in patients receiving Plasmalyte (4%) than it was in those receiving normal saline (2%) (Table 2). The SCOPE-DKA trial was a randomized controlled trial involving adult patients admitted to the intensive care unit with severe DKA at seven hospitals; these patients were randomized to receive either Plasmalyte or normal saline as the intravenous fluid management. The study showed a quicker resolution of DKA among patients receiving Plasmalyte than among those receiving normal saline. Their criteria to consider DKA resolution was a shift of base excess to more than or equal to –3 mEq/L at 48 hours. The study reported one death in a patient from the sodium chloride group due to mucormycosis.

In their cohort study within a randomized controlled trial, Attokaran et al. [14] reported a similar duration of hospital stay among patients receiving Plasmalyte-148 and normal saline after correcting for the diabetes type, pH, and blood glucose level (Tables 2, 4). However, compliance was low with assigned fluid in the Plasmalyte group, which might have underestimated the benefits of Plasmalyte. Both groups reported no mortalities. The percentages of patients admitted to the intensive care unit are shown in Table 4 [13,14].

As the studies by Ramanan et al. [11] and Attokaran et al. [14] were conducted in more than one healthcare facility, a strength of the studies is that their results may be generalizable to emergency departments and intensive care units at tertiary care centers. However, one limitation of the studies was the decreased compliance to study fluid in the Plasmalyte group compared to that in the normal saline group.

In their analysis of two RCTs, Self et al. [13] concluded that patients receiving balanced crystalloid solution had faster resolution of DKA (13.0 hours) than did those receiving normal saline (16.9 hours) (Table 3). Inpatient mortality associated with balanced crystalloid solution was lower (0 patients) than that with normal saline (one patient). Their criteria for DKA resolution was obtained from the American Diabetes Association Consensus Statement on Hyperglycemic Crisis, which includes a plasma glucose less than 200 mg/dL, and two of the following criteria: plasma bicarbonate ≥15 mEq/L, venous pH >7.3, or anion gap ≤12 mEq/L.

The strengths of the Self et al. [13] study include immediate assignment of patients to the sodium chloride versus balanced crystalloids group upon presentation to the emergency department, strong fluid group compliance and implementation of the study intervention into clinical practice. However, the study was limited by its single-center design, subgroup analysis of previous clinical trials and non-blinded nature.

Tsui et al. [15] compared the time to DKA resolution in adult intensive care unit patients receiving normal saline or balanced crystalloid (Ringer’s lactate or Plasmalyte); this group achieved a higher serum bicarbonate and pH with balanced crystalloids quicker than did those with sodium chloride. However, the study was limited by its statistically nonsignificant result of time to DKA resolution, small sample size, and preliminary study results.

In another retrospective study, DKA was observed to resolve at a comparable rate in patients who received normal saline (18.05 hours) and Plasmalyte (19.74 hours) [23]. The criteria for resolved DKA were adopted from the American Diabetes Association Consensus Statement on Hyperglycemic Crisis. Plasmalyte exhibited a greater increase in pH during the time frames of 4 to 6 hours and 6 to 12 hours.