Effectiveness of various balanced electrolyte solutions in the resuscitation of patients with diabetic ketoacidosis: A pairwise meta-analysis and network meta-analysis
In brief
Balanced fluids shortened diabetic ketoacidosis resolution by about 4 hours
Across 18 studies, balanced electrolyte fluids were linked to DKA resolving about 4.4 hours sooner than with normal saline, along with a modestly shorter hospital stay and improved bicarbonate and chloride levels. Benefits differed by fluid, adverse outcomes did not consistently improve, and indirect comparisons cannot establish which solution is best; head-to-head trials are still needed.
- Journal
- Diabetes research and clinical practice (Q1)
- Published
- 7 November 2026
- Study design
- Systematic review of cohort studies
- Evidence level
- Level 2, Moderate (CEBM 2a)
- Authors
- Vaishali Babasaheb Lote, Vikas S Sharma, Gunjan Vyas, Ashok K Shenoy
- PMID
- 42998723
- DOI
- 10.1016/j.diabres.2026.113645
Why clinicians should know about it
- Picked for Emergency Medicine (paper of the day, 9 November 2026): Comparative evidence for resuscitation fluids in diabetic ketoacidosis
- Picked for Internal Medicine (paper of the day, 9 November 2026): Fluid choice affects DKA resolution, length of stay, and AKI
Abstract
In diabetic ketoacidosis (DKA), choice of resuscitative fluid is debated, and superiority of balanced electrolyte solutions (BES) over each other remains unevaluated. This meta-analysis compared the effectiveness of BES with normal saline (NS) (18 studies), while network meta-analysis evaluated various BES (16 studies). PubMed, Cochrane Library, and EMBASE were searched from inception till April 28, 2025. Compared to NS, BES was associated with significantly shorter time to DKA resolution (MD = -4.41 h; 95 %CI = -7.00, -1.82), shorter length of hospital stay (LHS, MD = -0.43 days; 95 %CI = -0.72, -0.14), higher postresuscitation bicarbonate levels (PBL, MD = 1.20 mmol/L; 95 %CI = 0.48,1.92), and lower postresuscitation chloride levels (PCL, MD = -3.31 mmol/L; 95 %CI = -5.32, -1.29), though time to DKA resolution, PBL, and PCL had significant heterogeneity. Compared to NS, Sterofundin (MD = -4.30 h; 95 %CI = -6.14, -2.46; p < 0.0001) and RL (MD = -2.60 h; 95 %CI = -4.13, -1.07; p = 0.0009) had significantly shorter time to DKA resolution. Similarly, LHS was significantly shorter with Sterofundin (MD = -0.67 days; 95 %CI = -1.15, -0.19; p = 0.0066). PBL increased significantly with Plasmalyte (PL, MD = 2.16 mmol/L; 95 %CI = 1.12,3.20; p < 0.0001), and Ringer's lactate (RL, MD = 1.24 mmol/L; 95 %CI = 0.75,1.73; p < 0.0001), while PCL reduced significantly with RL (MD = -6.22 mmol/L; 95 %CI = -6.69, -5.76; p < 0.0001), Sterofundin (MD = -3.60 mmol/L; 95 %CI = -5.10, -2.10; p < 0.0001), and PL (MD = -3.16 mmol/L; 95 %CI = -4.58, -1.73; p < 0.0001). Rate of hyperkalemia (OR = 4.06; 95 %CI = 1.17,14.15) and acute kidney injury (OR = 2.26; 95 %CI = 1.00,5.08) was significantly higher with RL compared to NS. Thus, compared to NS, BES improved PBL and PCL, with faster time to DKA resolution and shorter LHS, but did not consistently improve adverse clinical outcomes. Comparison among BESs is indirect and hypothesis-generating; thus, direct head-to-head trials using patient-related outcomes are required in future trials.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.