Introduction and background

Diabetic ketoacidosis (DKA) is one of the most serious acute metabolic complications of diabetes mellitus, characterized by hyperglycemia, metabolic acidosis, and ketonemia. The global burden of DKA is substantial, with an estimated annual incidence of 4% to 8% among adults with type 1 diabetes mellitus. Furthermore, up to 80% of children and adolescents under 15 years of age present with DKA at the time of their initial type 1 diabetes diagnosis [1]. In industrialized nations, mortality rates range from 2% to 5%, whereas in developing countries, mortality can reach 6% to 24%, highlighting persistent gaps in acute diabetes care worldwide [2]. The cornerstone of DKA management encompasses three interdependent components: (1) aggressive intravenous fluid resuscitation to restore intravascular volume; (2) continuous intravenous insulin infusion to suppress ketogenesis and normalize blood glucose; and (3) meticulous electrolyte monitoring with appropriate replacement. Among these, fluid resuscitation is arguably the most critical initial intervention; however, the optimal choice of crystalloid solution for this purpose remains a subject of considerable clinical debate [2].

Historically, 0.9% normal saline (NS) has been the preferred fluid for DKA resuscitation, being a hypertonic solution with sodium and chloride concentrations of 154 mmol/L. This recommendation has been perpetuated by major clinical practice guidelines for decades [3]. However, the administration of large volumes of NS (often 4 to 8 liters over the first 24 hours of DKA treatment) carries the well-recognized iatrogenic risk of inducing hyperchloremic metabolic acidosis [1]. This occurs because the supraphysiological chloride load provided by NS exceeds the kidneys' capacity for chloride excretion, resulting in a non-anion gap metabolic acidosis that can superimpose upon the pre-existing elevated anion gap acidosis of DKA. Such hyperchloremia may obscure the true resolution of DKA, as clinicians may continue insulin infusion based on a persistently low bicarbonate level that is actually driven by chloride rather than ongoing ketogenesis. Furthermore, hyperchloremia has been associated with renal vasoconstriction, reduced glomerular filtration rate, and a higher risk of acute kidney injury (AKI) in critically ill patients, raising legitimate concerns about saline's safety as the standard resuscitation fluid in DKA [4].

Balanced crystalloid (BCs) solutions, such as Ringer's lactate (RL), Plasma-Lyte 148, Isolene, and Sterofundin, have surfaced as potentially beneficial substitutes to deal with these problems. BCs are designed with electrolyte compositions more closely approximating that of human plasma, typically containing chloride concentrations between 98 and 113 mmol/L, which is substantially lower than that of NS. The theoretical advantages of BCs in DKA are compelling: by avoiding an excessive chloride load, they may prevent or attenuate the development of iatrogenic hyperchloremic acidosis, thereby allowing the anion gap to serve as a more reliable marker of true metabolic resolution. Additionally, BCs may limit the risk of AKI, shorten the duration of insulin infusion, reduce the amount of fluid needed for resuscitation, and shorten the length of stay (LOS) in the intensive care unit [5]. However, lingering concerns regarding the lactate content of some balanced solutions (typically 28 mmol/L in lactated Ringer's (LR)) and the theoretical risk of exacerbating lactic acidosis in hypoperfused patients have historically limited their widespread adoption in DKA, despite a lack of robust evidence supporting such concerns.

The goal of this systematic review was to assess the clinical outcomes of fluid resuscitation in patients with DKA using BCs versus plain saline, given the ongoing uncertainty regarding the best fluid for DKA resuscitation and the growing but contradictory data from recent trials.