Fluids and Electrolytes

Dr. Steven Alexander's Fluid Primer

FEN Basics Powerpoint, Kevin Kuo

Fluid Resuscitation

  • Bolus of fluids defined as 20 cc/kg given over 5-10 minutes (Surviving Sepsis Guidelines). Hence, do not program 999 ml/hr on the pump. Pull and push directly or infuse via pressure bag.
  • No difference in outcomes whether one uses albumin vs. normal saline (SAFE Study, NEJM 2004)
    • Similar findings with other trials such as VISEP (10% pentastarch vs. LR), 6S (6% HES vs. Ringer's acetate), and CHEST (HES vs. NS) which generally showed worsened renal outcomes in the starch groups
  • Some recent evidence suggesting treatment with balanced fluids (lactated Ringer's) may be preferable to normal saline with associated improvements in outcome (Raghunathan, CCM 2014). Other studies (Stenson et al, PCCM 2018) have shown hyperchloremia (i.e. from normal saline boluses) is independently associated with worse outcomes in pediatric sepsis and improved outcomes with the use of balanced fluids in pediatric sepsis (Emrath et al, CCM 2017). Other retrospective studies have not shown a difference (Weiss et al, J Ped 2016). 
  • The SMART trial also demonstrated improved composite outcomes with balanced solutions vs. saline in critically ill adults (Semler et al, NEJM 2018). Most recently, BASICS (JAMA 2021), a RCT of 10500 adult critically ill patients in Brazil showed no difference in mortality between those randomized to a balanced solution vs. saline. 
  • Reasonable to limit overagressive resuscitation of patients with DKA as there may be some association of increased fluid administration with increased risk of cerebral edema- i.e. no fluid bolus or 10 cc/kg fluid bolus only unless other evidence of decompensated shock

Figure 1: Electrolyte Composition of Commonly Used Intravenous Fluids

Figure 2: Electrolyte makeup of common bodily fluids

Diuresis and Fluid Removal

How I think about diuresis in the PICU:

1) Why am I diuresing this patient? Typically it is to improve pulmonary edema, pleural effusions, lung compliance, etc and to facilitate improvements in respiratory status and deescalation of support (ventilator settings, coming off mechanical ventilation, etc). Basically, dry lungs are happy lungs. Diuresis can also sometimes help with intraabdominal pressure (ie ascites), heart failure (leading to pulmonary or hepatic congestion), etc.

2) Set a fluid goal for the day. How much do you want to take off? A general rule I use is 20 cc/kg/day as a baseline for moderate diuresis. The circulating blood volume for most of our patients is 70-80 cc/kg and so this will take off about 25% of the circulating blood volume. Remember that your goal is to remove interstitial or "third-space" fluid, not actually intravascular volume but this is the only way we can get at that fluid, by reducing intravascular hydrostatic pressure. This is described by the Starling equation, which describes net filtration out of the capillary as a balance between hydrostatic and oncotic forces. The reflection coefficient (sigma) describes the propensity of a substance to stay in the vasculature (for example, sodium is 1.0, meaning it stays within the vasculature perfectly, vs. mannitol is closer to 0.75-0.9, meaning it eventually leaks out of the vessels to some degree, pulling water with it back to the interstitium. 

General rule: ~ 20 cc/kg/day
If you are relatively hypotensive, have concerns for worsening AKI, want to be more ginger, etc: 10 cc/kg/day
If you are relatively hypertensive, are really up against a wall from a respiratory standpoint with high settings etc: 40 cc/kg/day

These are general guidelines that can then be adjusted based on how the patient is doing (lung compliance, blood pressure, renal function, etc)

Starling Equation describing net filtration out of the capillary (by convention) as determined by hydrostatic and oncotic pressures

3) You've chosen a fluid goal. Now choose a starting diuretic regimen. What you choose (ie lasix 1 mg/kg IV q8h)  is much less important than how you monitor and adjust throughout the day to achieve your goal. Typical starting regimens and other considerations include:

    a) Lasix 0.5 mg/kg IV or PO (IV thought to be twice as potent) either as a single spot dose up to as frequently as q6 hours (lasix- "lasts six" hours). In adult sized patients, smaller doses such as 10-20 mg may be sufficient in a lasix naive patient. 1 mg/kg may be more appropriate for patients who are not lasix naive. 

    b) Consideration of addition of Diuril (chlorthiazide) to augment diuresis. The distal convoluted tubule will try to counteract the loop diuretic, lasix, you've given, by reabsorbing sodium. Diuril poisons the sodium channel's reabsorption, leading to a synergistic effect on diuresis. This can be dose 5 mg/kg up to q6 hours given shortly after the lasix (in order to have that synergystic effect). In some circumstances, double dose diuril at 10 mg/kg is utilized as well.

    c) Consideration of a lasix infusion- this can be helpful to have a steady dose of diuretics to avoid swings with higher drug levels and thus "dumping" of urine which can occasionally have hemodynamic effects like hypotension. Lasix infusions are dosed mg/kg/hr. Hence, 0.2 mg/kg/hr would be the equivalent of 1.2 mg/kg IV q6 (a relatively moderate/high dose). A typical starting dose might be 0.05-0.15 mg/kg/hr which would be equivalent to 0.3 mg/kg-0.9 mg/kg IV q6 and then adjusted based on response. 

    d) Consider how often you need to check electrolytes (primarily potassium as well as monitoring renal function). With aggressive diuresis, I typically plan to check at least q12h if not more frequently. 

    e) Consider 25% albumin. As seen by the Starling equation, oncotic pressure also helps determine how much fluid moves out from the interstitium back into the vasculature. At serum albumin levels <2.5 (or occasionally <3), one could consider 0.5 mg/kg of 25% albumin (max 25 g) IVq6 for 4 doses to run for several hours before each dose of lasix/diuril in order to "pull fluid" from the interstitium and then remove it via the diuretics. Exogenous albumin will eventually leak into the interstitium (ie radiolabeled albumin can be seen to do so) and thus it is only potentially effective to augment diuresis. Evidence for this approach is limited but experientially, does seem to augment diuresis. 

    f) If your response to diuretics is poor, consider how your renal function is, ensure you have a sufficient level of chloride (typically want >90 as lasix works on the Na K 2Cl channel) , or consider an alternative agent like bumetanide, which although is in the same class, we sometimes see a different effect. 

4) Again, MONITOR your patient's fluid status and adjust throughout the day to meet the predetermined fluid goal from A. This is key. 

5) Sometimes it is worth diuresing until you see a little hypotension and even accepting that and using a vasoactive agent to facilitate further diuresis. Similarly, sometimes with bad lung disease, we will diurese until we see biochemical evidence of some intravascular depletion and potentially even AKI (BUN elevated, bump in creatinine etc), recognizing we are making a conscious choice to lean on the kidneys to help the lungs. 

6) Once you've achieved the goal of your diuresis (ie get off the ventilator etc), reassess whether you still need the same (or any) level of diuresis/fluid removal. 

The nephron and site of action for various diuretics

Sodium Disorders

  • Common in the pediatric intensive care unit and hospitalized patients in general
  • Hyponatremia generally occurs as a result of increased ADH secretion (secondary to pulmonary or CNS pathology, pain, narcotic administration, etc) which promotes free water retention
  • This is the rationale for checking frequent sodiums on postoperative patients who have had a neurosurgical procedure such as a craniotomy or craniosynostosis repair
  • Due to the propensity of hospitalized patients to have increased ADH and free water retention, most patients in the PICU (with the exception of neonates <28 days, specific requests by specialty services, or in the case of unintentional iatrogenic hypernatremia) can be placed on isotonic (ie D5 normal saline) fluids as maintenance (vs. D5 1/2 normal saline). A RCT of 690 hospitalized pediatric patients (McNab et al, Lancet 2014) demonstrated a lower rate of hyponatremia (4 vs 11% p=0.001) with the use of isotonic saline vs. 1/2 normal saline. There were no significant differences in the rate of adverse events. The authors conclude isotonic saline should be used as maintenance fluid for hospitalized pediatric patients.
  • Can also occur due to cerebral salt wasting (unclear pathophysiology but theory that neurologic injury leads to natriuretic peptide release that promotes loss of sodium and fluid). Treatment includes fluid/sodium repletion and potentially fludrocortisone
  • Workup of sodium disorders includes evaluation of fluid status, serum Na, urine output, and urine osmolality and can be differentiated by the table below:
Figure 3: Expected Changes for Common Sodium Disorders
  • Immediate life threatening hyponatremia (ie seizures, comatose) can be treated with hypertonic saline (3%) titrated to effect 
  • Note there is a risk of central pontine myelinolysis with rapid correction of longstanding hyponatremia as fluid rushes out of neurons and into the intravascular space
  • Conversely, there is a risk of cerebral edema with overly rapid correction of chronic hypernatremia
  • In general, the recommendation is to correct chronic disorders of sodium no faster than 0.5-1 mEq/hr or a total of 12 mEq/L/day to avoid rapid shifts in tonicity. Careful and frequent monitoring with adjustment is critical.
  • Change in serum Na= Infusate Na -Serum Na/(Total body water +1) where TBW= 0.6(weight in kg) 
NEJM Review Article (Adrogue NEJM 2000) 
  • Response in sodium to fluid depends on the underlying pathology and not just the amount of sodium in the fluid administered. For example, if a patient has SIADH, a urine osm of 600 and a serum Na of 125, giving them normal saline (Sodium of 154 mEq/L) will paradoxically decrease their serum Na and make the hyponatremia worse. This is because normal saline has an osm of ~300 and so the patient can excrete the 300 osm of solute you gave in 1/2 L of urine, leaving essentially 1/2 L of free water
  • Other causes of hyponatremia include adrenal insufficiency (generally accompanied by hyperkalemia), decreased effective circulating volume leading to appropriate ADH secretion (ie CHF, nephrotic syndrome), or hypothyroidism
  • 3% hypertonic saline given in a dose of 5 cc/kg is expected to raise the serum sodium approximately 5 mEq/L. 

Potassium Disorders

  • Hypokalemia frequently occurs as a result of diuretics. Beta agonists, insulin, and alkalemia also promote entry of potassium into cells and thus relative serum hypokalemia. ECG changes include U waves or t wave flattening/inversion
  • Hypokalemia can lead to arrhythmias as well as skeletal muscle weakness
  • Significant risk of iatrogenic injury with overshoot hyperkalemia when treating hypokalemia
  • Hyperkalemia occurs with hemolysis, massive cell injury, acidemia, aldosterone deficiency (or spironolactone use), and renal failure 
  • Hyperkalemia can lead to life threatening ECG changes and arrhythmias including peaked T waves, bradycardia, wide complex QRS complexes, sine wave pattern, and ventricular arrhythmias
  • Treatment of hyperkalemia includes sodium polystyrene sulfonate (kayexlate) to bind potassium (more chronic) (note, newer  cation exchangers-ie, patiromer or zirconium cyclosilicate- are likely more effective if available) Calcium chloride (20mg/kg) in a central IV to stabilize the cardiac membrane or calcium gluconate 100 mg/kg IV in a peripheral IV, insulin and dextrose (0.2 units/g glcuose and 1 g/kg glucose), sodium bicarbonate (1 mEq/kg IV), fuorsemide, albuterol, and renal replacement therapy

Figure 4: EKG changes in hypo and hyperkalemia

Magnesium Disorders

  • Hypomagnesemia typically occurs in the PICU due to loop diuretics or transplant immunosupressives (i.e. tacrolimus or cyclosporine)
  • Hypomagnesemia is associated with ventricular arrhythmias, torsades de pointes, seizures, tetany, fasciculations, and coma. Repletion 25-50 mg/kg magnesium sulfate IV over 15-60 minutes
  • Hypermagnesemia occurs with renal failure or via iatrogenic administration. It is associated with decreased CNS responsiveness, depressed DTR's, and hypotension. IV calcium can be used to treat life threatening hypermagnesemia.

Phosphorous Disorders

  • Hypophosphatemia generally associated with phsophate deficient TPN, refeeding syndrome, DKA, or severe respiratory alklalosis
  • Low phosphorous can lead to impaired energy utilization and diaphragmatic and respiratory muscle weakness, neuropathy and weakness, impaired cardiac output, tissue hypoxia due to reduced levels of 2-3 DPG, and impaired immune responses due to decreased available ATP. Generally treated with IV phosphorous repletion in the PICU (0.15 to 0.3 mmol/kg/dose)
  • Hyperphosphatemia generally associated with renal failure although phosphate containing enemas have also been reported as causes of hyperphosphatemia
  • Hyperhphosphatemia can be treated with sevelamer to beind phosphorous, mannitol diuresis, and renal replacement therapies

Calcium Disorders

Figure 5: Calcium homeostasis

  • Hormonally regulated by PTH, viatmin D, and calcitonin
  • PTH inhibits phosphorous reabsorption and raises serum calcium levels by shifting it from bone to the ECF
  • Vitamin D stimulates intestinal absorption of calcium and phosphorous
  • Calcitonin is produced by c cells of the thyroid in response to elevated ionized Ca levels and leads to lower serum Ca levels
  • Hypocalcemia can be caused by PTH deficiency, vitamin D deficiency, hypercalcitoninemia, magnesium depletion, or iatrogenic causes such as significant blood product administration (citrate anticoagulant), or excessive citrate during renal replacement therapy
  • Hypocalcemia leads to tetany (Chovstek and Trousseau signs), seizures, hypotension
  • Treatment of hypocalcemia involves treating the underlying cause as well as oral/IV supplemenation (100 mg/kg calcium gluconate or 20 mg/kg calcium chloride)
  • Hypercalcemia can result from hyperparathyroidism, bone lysis, vitamin D intoxication
  • Hypercalcemia can lead to changes in CNS function, decreased QT interval, and impaired nerve conduction
  • Hypercalcemia is treated with fluid hydration and furosemide diuresis to promote calciuresis. Thiazides promote calcium retention and should not be used. Calcitonin has also been used to promote calciuria and prevent bone resporption

Electrolyte Derangement Chart

Courtesy of Richard Pierce, MD


1) Finfer S, Bellomo R, Boyce N, French J, Myburgh J, Norton R; SAFE StudyInvestigators. A comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med. 2004 May 27;350(22):2247-56. PubMed PMID: 15163774.

2)Raghunathan K, Shaw A, Nathanson B, Stürmer T, Brookhart A, Stefan MS,Setoguchi S, Beadles C, Lindenauer PK. Association between the choice of IV crystalloid and in-hospital mortality among critically ill adults with sepsis*. Crit Care Med. 2014 Jul;42(7):1585-91.

3) Au AK, Ray PE, McBryde KD, Newman KD, Weinstein SL, Bell MJ. Incidence ofpostoperative hyponatremia and complications in critically-ill children treated with hypotonic and normotonic solutions. J Pediatr. 2008 Jan;152(1):33-8.

4) Montañana PA, Modesto i Alapont V, Ocón AP, López PO, López Prats JL, ToledoParreño JD. The use of isotonic fluid as maintenance therapy prevents iatrogenic hyponatremia in pediatrics: a randomized, controlled open study. Pediatr Crit Care Med. 2008 Nov;9(6):589-97.

5) Finberg L. Hypernatremic (hypertonic) dehydration in infants. N Engl J Med.1973 Jul 26;289(4):196-8.

6) Gennari FJ. Hypokalemia. N Engl J Med. 1998 Aug 13;339(7):451-8.

7) Fiser RT, Torres A Jr, Butch AW, Valentine JL. Ionized magnesiumconcentrations in critically ill children. Crit Care Med. 1998 Dec;26(12):2048-52.

8)  140 mmol/L of sodium versus 77 mmol/L of sodium in maintenance intravenous fluid therapy for children in hospital (PIMS): a randomised controlled double-blind trial
McNab, Sarah et al. The Lancet , Volume 385 , Issue 9974 , 1190 - 1197