Risk of Electrolyte Disorders in Acutely Ill Children Receiving Plasmalike Isotonic Fluids

Study Overview

• Randomized clinical pragmatic trial (unblinded) conducted at the Pediatric Emergency Department (ED) of Oulu University Hospital, Finland.
• Time frame: October 3, 2016 – April 15, 2019.
• ClinicalTrials.gov identifier: NCT02926989; EUDRA-CT 2016-002046.
• Compared two commercially available intravenous (IV) maintenance solutions in acutely ill children:
  • Plasmalike isotonic fluid: 140 mmol/L Na⁺, 5 mmol/L K⁺, 1.5 mmol/L Mg²⁺, 98 mmol/L Cl⁻, 23 mmol/L acetate, 23 mmol/L gluconate, 5 % dextrose.
  • Moderately hypotonic fluid (control): 80 mmol/L Na⁺, 20 mmol/L K⁺ in 5 % dextrose (historical textbook recommendation).
• Focus: Risk of electrolyte disorders (hypokalemia, hypernatremia, hyponatremia) and fluid retention.

Background & Rationale

• Since 2018, the American Academy of Pediatrics (AAP) recommends isotonic maintenance fluids to avoid hyponatremia.
• Prior fatal cases of postoperative hyponatremia linked to hypotonic IV fluids (Refs 2-6).
• Evidence gaps noted by AAP:
  • Limited data on hypernatremia and fluid retention with isotonic therapy.
  • Sparse reporting of hypokalemia when potassium content is low (5 mmol/L) in plasmalike solutions.
• Adult volunteer data: Isotonic fluids ↓ renal output (possible fluid retention).
• Pediatric trials have largely focused on ICU/post-operative settings, not general ED populations.

Research Objectives

• Primary: Determine proportion of children developing any clinically significant electrolyte disorder during hospitalization:
  • \text{Hypokalemia} < 3.5\ \text{mmol/L}
  • \text{Hyponatremia} < 132\ \text{mmol/L}
  • \text{Hypernatremia} > 148\ \text{mmol/L}
• Main secondary: Incidence of severe hypokalemia (<3.0\ \text{mmol/L}) and weight change (fluid retention proxy).
• Additional prespecified secondary metrics: mild hyponatremia, need to modify fluids, ICU transfer, IV duration, length of stay, 30-day mortality.
• Exploratory post-hoc: time-to-disorder, metabolic acidosis resolution, copeptin (ADH surrogate) at 6-24 h.

Study Design & Methods

• Design: Pragmatic, open-label, intention-to-treat; clinicians could adjust therapy for safety (reflects real-world practice).
• Randomization: 1:1 using permuted blocks of 4; allocation via opaque sealed envelopes.
• Inclusion criteria: 6 mo – 12 y, acute illness requiring hospitalization + IV fluids.
• Key exclusions:
  • Plasma Na⁺
  • Need for 10 % glucose, diabetes (types 1, ketoacidosis, insipidus), renal replacement therapy, severe liver disease, oncology hydration protocols, inborn metabolic errors.
• Fluid rate: Holliday-Segar formula for maintenance \bigl(100\ \text{mL/kg for 1st 10 kg} + 50\ \text{mL/kg for 2nd 10 kg} + 20\ \text{mL/kg for each kg >20}\bigr) plus rehydration bolus (5-10 % body weight).
• Monitoring:
  • Electrolytes daily (more if indicated); open-label results to clinicians.
  • Weight before start, daily, and post-therapy.
  • Blood gases as clinically indicated (venous preferred; capillary if needed).

Participant Characteristics (Intention-to-Treat = 614)

• Mean age $4.0\ \text{y} \ (\text{SD } 3.1)$.
• Sex: 51 % boys (315/614).
• Language: 100 % Finnish speaking.
• Common admission diagnoses:
  • Respiratory infections (≈32-35 %).
  • Gastroenteritis (≈26-27 %).
  • Pneumonia, viral wheezing, pyelonephritis, sepsis, surgical illnesses (appendicitis, empyema, etc.).
• Dehydration present in ~1/3; 25 % received pre-randomization fluid bolus.
• ICU admission from ED: 1.5 % (9/614).
• Baseline labs: $\text{Na}^+ \approx 138\ \text{mmol/L};\ \text{K}^+ \approx 4.1\ \text{mmol/L}$, pH $\approx 7.40$.

Sample Size & Power

• Assumed electrolyte disorder rate in isotonic group = 13 %.
• Detect absolute risk reduction of 7 % with 80 % power, $\alpha = 0.05$ → 275 per arm; targeted 305/group to offset missing data.

Primary Outcome Results

• Electrolyte disorders:
  • Isotonic: 61/308 (20 %).
  • Hypotonic: 9/306 (2.9 %).
  • Absolute difference: 17 % (95 % CI 12-22 %).
  • $RR = 6.7\ (95\%\ CI\ 3.5{-}13)$.
  • Number Needed to Harm (NNH) $= 6\ (95\%\ CI\ 5{-}9)$ for isotonic fluid.

Breakdown

• Hypokalemia (
• Hypernatremia (>148 mmol/L):
  • 4 vs 0 (1.3 % vs 0 %); absolute diff = 1.3 %, P = .04.
• Hyponatremia (

Secondary Outcomes

• Severe hypokalemia (
• Mild hyponatremia (132-135 mmol/L): no significant difference (2.3 % vs 3.6 %).
• Weight gain (fluid retention proxy):
  • Mean ± SD: 279 ± 431 g (isotonic) vs 195 ± 420 g (hypotonic).
  • Mean difference: 84 g (95 % CI 16-154 g); P = .02 (statistically, but arguably minimal clinical impact).
• No significant differences in:
  • Fluid therapy modifications (~13-15 % in both groups).
  • ICU transfers after admission (≈3 %).
  • Duration of IV therapy (~29 h) or total hospitalization (~2.3 days).
  • Mortality: 0 deaths (30-day window).

Exploratory Post-Hoc Findings

• Time to disorder:
  • Hypernatremia developed by 12 ± 4.9 h post-start in isotonic group.
  • Hypokalemia onset: 14 ± 8.4 h (isotonic) vs 31 ± 26 h (hypotonic); P ≈ .07.
• Metabolic acidosis resolution (blood gas subgroup):
  • Day-1 bicarbonate <21 mmol/L: 5.3 % (isotonic) vs 26 % (hypotonic); absolute diff = –21 % (benefit from isotonic sodium bicarb precursors acetate/gluconate).
  • Base excess <−2.5: 11 % (isotonic) vs 33 % (hypotonic).
• Copeptin (10 % random sample): higher mean in isotonic (8.1 pmol/L) vs hypotonic (7.3 pmol/L) but NS.

Severe Adverse Events & Safety Notes

• No neurologic complications attributed to dysnatremia.
• One isotonic-arm child (congenital nephrosis + adenovirus GE) transferred to PICU for severe hypokalemia $K^+ = 2.2\ \text{mmol/L}$.
• Another isotonic patient developed seizures; later diagnosed with epilepsy (association uncertain).

Clinical Implications & Recommendations

• Commercial plasmalike isotonic solutions with only 5 mmol/L K⁺ substantially increase hypokalemia risk; potassium supplementation should be routine when isotonic Na⁺ ≈140 mmol/L is used.
• Hypernatremia, though infrequent, occurs only in isotonic arm → requires surveillance, especially during first 12 h.
• Moderate hypotonic fluid containing both sodium 80 mmol/L and potassium 20 mmol/L appears safe in non-ICU acutely ill children, contradicting fear of hyponatremia in this population.
• Weight gain difference small but supports vigilance for sodium-related fluid retention in vulnerable patients (e.g., cardiac, renal, neuro edema risk).
• Findings challenge blanket AAP recommendation favoring isotonic fluids for all children; nuance needed based on clinical scenario, potassium content, and monitoring capacity.

Connections to Previous Literature

• Echoes earlier ICU/post-operative trials showing isotonic ↓ hyponatremia but extends knowledge by:
  • Demonstrating hypokalemia & hypernatremia risks in ED admissions.
  • Highlighting deficiency of potassium in plasmalike products.
• Contrasts with meta-analysis wide CI (RR 0.6-2.4) for hypernatremia—this trial narrows evidence towards harm.
• Supports adult volunteer data on isotonic fluid–induced renal water retention (weight gain signal).

Ethical, Philosophical & Practical Considerations

• Pragmatic open design respects clinician judgment, improving external validity but introduces potential bias in discharge timing.
• Informed consent from guardians & assent (6-12 y) uphold autonomy.
• Highlights equity issue: Commercial fluid compositions are not one-size-fits-all; reliance on ready-made products without tailoring electrolytes can harm.
• Encourages healthcare systems to stock customizable IV additives or higher-potassium isotonic options.

Limitations

• Single-center Finnish cohort → cultural/genetic/environmental factors may limit global generalizability.
• Excluded renal-impaired, diabetics, severe electrolyte derangements; results not applicable to those high-risk groups.
• Open-label may influence clinician behavior (e.g., earlier discharge, fluid adjustments), diluting treatment effect on hyponatremia.
• Mortality & rare neurologic outcomes under-powered.
• Product-specific: Plasmalyte Glucos composition may differ from other isotonic solutions (e.g., balanced crystalloids with 4 mmol K⁺ or none).

Key Numerical & Statistical Data (LaTeX)

• $RR_{electrolyte\ disorder} = 6.7\ (95\%\ CI\ 3.5{-}13)$
• $NNH_{electrolyte\ disorder} = 6\ (95\%\ CI\ 5{-}9)$
• $RR_{hypokalemia} = 6.3\ (95\%\ CI\ 3.2{-}12)$
• $RR_{severe\ hypokalemia} = 7.9\ (95\%\ CI\ 1.3{-}49)$
• $\Delta Weight = 84\ \text{g}\ (95\%\ CI\ 16{-}154)$
• Sample size formula (simplified): $n = \frac{(Z<em>{1-\alpha/2}+Z</em>{1-\beta})^2\bigl[p<em>1(1-p</em>1)+p<em>2(1-p</em>2)\bigr]}{(p<em>1-p</em>2)^2}$ where $p<em>1=0.13,\ p</em>2=0.06$ leading to ~275/group.

Practical Take-Home Messages

• Do not assume plasmalike isotonic fluids are universally safer; consider potassium concentration.
• For acutely ill, non-ICU pediatric patients with likely pre-existing hypotonic losses, moderately hypotonic (Na⁺ 80, K⁺ 20) may be optimal.
• Monitor electrolytes within the first 12-24 h; most disorders emerged early.
• Corrective actions: add KCl to isotonic fluids or switch fluid type per labs.
• Future research: multicenter trials including renal-impaired children, exploration of isotonic fluids premixed with 20 mmol/L K⁺, long-term neurocognitive outcomes of transient electrolyte shifts.