Every bolus you write goes to a compartment. Pick the wrong fluid and you shift water the wrong way — dropping sodium in SIADH, expanding the third space in sepsis, or chasing your own tail in DKA. Work through the compartments first, then open the bolus simulator.
Total body water ≈ 60% of lean body mass. Two-thirds of that water sits inside cells (ICV); one-third is extracellular (ECV). The ECV itself splits 3:1 — three parts interstitial, one part intravascular. Every fluid you order lands somewhere in this map.
For any euvolemic adult, TBW ≈ 60% of lean mass (women ~55%, elderly ~50%, children ~65%). Of that, 2/3 is intracellular, 1/3 extracellular; the ECV splits 3:1 interstitial to plasma. Drag the sliders — every label on the map below updates live.
At equilibrium, osmolality is equal in ICV and ECV (~285 mOsm/kg). Water moves freely across cell membranes; solute (Na⁺, K⁺) does not. So tonicity sets the ICV/ECV split — and then within the ECV a second rule (Starling forces) sets how much stays in plasma vs interstitium:
Corollary — and the most-missed point on rounds: "isotonic stays in the ECV" is not the same as "isotonic stays intravascular." Only about a quarter of any isotonic crystalloid hangs around in plasma — which is why septic-shock resuscitation needs such big volumes, and why aggressive NS runs show up later as tissue edema, not just BP support.
The other corollary: you can shrink a cell with 3% saline, swell it with D5W, and leave ICV untouched with NS.
Every bag on the shelf fits one of five boxes. Tonicity (vs plasma ~285 mOsm) and whether it contains an oncotic agent (albumin, starch) are the only two axes you need.
0.9% NaCl · Lactated Ringer's · Plasma-Lyte
1 L bolus → 0 mL ICV + 750 mL ISF + 250 mL plasma (Starling 3:1). The plasma fraction raises BP; the interstitial fraction is the edema you'll see tomorrow.
D5W · 0.45% NaCl · D5 ¼NS
1 L D5W → 667 mL ICV + 250 mL ISF + 83 mL plasma. Dangerous in SIADH (drops Na further).
3% NaCl · 23.4% NaCl · Mannitol
1 L 3% NaCl → ICV gives up ~2.7 L → ECV gains ~3.7 L (2.7 L ISF + 0.9 L plasma). Shrinks cells. Use for symptomatic hyponatremia or raised ICP.
5% Albumin · 6% HES (avoided) · FFP
1 L 5% albumin → 0 mL ICV + ~100 mL ISF + ~900 mL plasma. Expensive but efficient when IVF must rise without crystalloid overload (SBP, hepatorenal).
25% Albumin
100 mL 25% albumin → ICV unchanged · ISF gives up ~300 mL · plasma gains ~400 mL. Useful when ISF is loaded (cirrhosis) and IVF is empty.
LR · Plasma-Lyte A
1 L LR → ~50 mL ICV + ~710 mL ISF + ~240 mL plasma (near-isotonic but slightly hypotonic vs plasma). Lower Cl⁻ and buffered — less hyperchloremic acidosis and AKI vs NS (SMART, SALT-ED, PLUS 2022).
Every fluid you can order, side-by-side. Compare Na, Cl, buffer, and osmolality against plasma (Na 140, Cl 103, osm 285).
| Fluid | Tonicity | Na⁺ | Cl⁻ | K⁺ | Ca²⁺ | Buffer | Glucose | Osm | pH |
|---|---|---|---|---|---|---|---|---|---|
| Plasma | Reference | 140 | 103 | 4 | 2.5 | HCO₃ 24 | 5 | 285 | 7.40 |
| 0.9% NaCl (NS) | Iso | 154 | 154 | 0 | 0 | — | 0 | 308 | 5.5 |
| Lactated Ringer's | Balanced | 130 | 109 | 4 | 3 | Lactate 28 | 0 | 273 | 6.5 |
| Plasma-Lyte A | Balanced | 140 | 98 | 5 | 0 | Acet 27 · Gluc 23 | 0 | 294 | 7.4 |
| 0.45% NaCl (½NS) | Hypo | 77 | 77 | 0 | 0 | — | 0 | 154 | 5.5 |
| D5W | Hypo* | 0 | 0 | 0 | 0 | — | 50 g/L | 252 | 4.0 |
| D5 ½NS | Hypo* | 77 | 77 | 0 | 0 | — | 50 g/L | 406 | 4.5 |
| D5 NS | Iso* | 154 | 154 | 0 | 0 | — | 50 g/L | 560 | 4.0 |
| 3% NaCl | Hyper | 513 | 513 | 0 | 0 | — | 0 | 1026 | 5.0 |
| 23.4% NaCl | Hyper | 4000 | 4000 | 0 | 0 | — | 0 | 8008 | 5.0 |
| 5% Albumin | Iso-oncotic | 145 | 145 | 0 | 0 | Albumin 50 g/L | 0 | ~300 | 7.0 |
| 25% Albumin | Hyper-oncotic | 145 | 145 | 0 | 0 | Albumin 250 g/L | 0 | ~1500 | 7.0 |
* Dextrose is rapidly metabolized, so effective (in vivo) tonicity is set by the saline content, not the measured osm. D5W is effectively free water after glucose metabolism.
Pick a patient physiology, a fluid, and a volume. The model honors scenario-specific physiology — capillary leak in sepsis, third-spacing in cirrhosis, ADH lock in SIADH — not just fluid composition.
~250 mL plasma · ~750 mL interstitium · 0 mL intracellular
~660 mL intracellular · ~250 mL interstitium · ~90 mL plasma
Light band = volume before bolus. Accent band = volume after equilibration. Watch where the water lands.
Click a preset to load a realistic case into the simulator.
Scenario → right fluid → why. Take any of these into the simulator to watch the mechanism play out.
Pick: Lactated Ringer's or Plasma-Lyte, 30 mL/kg over first 3 h.
Why: SMART (2018, n=15,802) showed balanced crystalloid reduced major adverse kidney events (MAKE-30) vs NS — HR 0.91. PLUS (2022) was neutral but confirmed safety. Avoid NS-only high-volume resuscitation; hyperchloremic acidosis worsens AKI.
Bonus: In septic shock that remains hypotensive after crystalloid, add 5% albumin (ALBIOS 2014). Avoid HES (6S, CHEST — worse renal outcomes).
Pick: Start with NS 15–20 mL/kg × 1 h. After corrected Na > 135, switch to ½NS or D5½NS when glucose < 250.
Why: The initial deficit is a profound extracellular volume loss (osmotic diuresis). NS restores IVF. The switch to hypotonic is deliberate once Na stabilizes: DKA patients have a free water deficit masked by hyperglycemia.
PLUS (2022) showed LR vs NS equivalent in DKA with possibly faster DKA resolution on LR. Some centers have moved to LR first-line.
Pick: 3% NaCl 100–150 mL IV bolus × 1–3. Target: Na up 4–6 mEq/L fast, then stop.
Why: Only hypertonic saline rapidly reverses cerebral edema. Each 100 mL bolus raises Na by ~2 mEq/L in a 70 kg adult. Stop at symptomatic improvement; the first 4–6 mEq/L is life-saving, the next 6 is dangerous (ODS).
See the acute hyponatremia journey for the full protocol.
Pick: 25% albumin 1 g/kg day 1, 20–40 g/day thereafter (alongside terlipressin).
Why: Cirrhotic patients are interstitially overloaded but intravascularly empty. Hyper-oncotic albumin pulls water from the loaded ISF into plasma. Pure crystalloid worsens ascites without fixing IVF.
Also proven for SBP (1.5 g/kg day 1, 1 g/kg day 3 — halves mortality) and large-volume paracentesis (>5 L tapped → 6–8 g albumin per L removed).
Pick: 3% NaCl (boluses of 250 mL) or 23.4% NaCl (30 mL central line push) for acute herniation. Maintenance: NS (never hypotonic).
Why: Hyperosmolar therapy pulls water from brain cells and lowers ICP within minutes. Mannitol works similarly but causes osmotic diuresis → paradoxical volume loss. Hypertonic preserves volume status.
LR and Plasma-Lyte are slightly hypotonic (273–294 mOsm) and can modestly worsen cerebral edema — use NS in the TBI patient.
Pick: D5 ½NS + 20 mEq KCl/L at ~1.5 mL/kg/h for a healthy adult. Or isotonic (LR) if any risk of SIADH.
Why: The 4-2-1 rule (100/50/20 mL/kg/day → 4/2/1 mL/kg/h) gives ~2000 mL/day for a 70 kg adult. Free water component covers insensible losses. Dextrose covers ~200 kcal and prevents ketosis.
Recent shift: Many centers now default to balanced isotonic maintenance to reduce iatrogenic hyponatremia risk in hospitalized patients with elevated ADH (surgery, pain, nausea — all potent ADH stimuli). PADS 2018 meta showed 17% absolute risk reduction in hyponatremia with isotonic.
Mistakes that show up on rounds. Memorize the reasoning, not just the rule.
SIADH patients cannot excrete free water. D5W is essentially pure free water after metabolism. Giving 1 L D5W to a patient already at Na 128 will drop Na by 2–4 mEq/L within hours — not raise volume status. In SIADH, the treatment is restriction, salt, or vaptan, never free water.
A hyponatremic patient who "looks dry" and gets repeated NS boluses may stay dry AND drop Na. NS delivers 154 mEq Na/L — if serum Na is 120, each liter raises Na by only ~0.3 mEq/L and may briefly suppress ADH enough to unmask a dangerous free-water diuresis (the "desalination" phenomenon — Steele 1997). Monitor Na q2–4 h during resuscitation of severe hyponatremia.
In sepsis with capillary leak, every 1 L of NS bolus puts ~200–300 mL into plasma, ~700 mL into a leaky ISF, and contributes to tissue edema that delays wound healing, impairs oxygen diffusion, and ultimately costs ICU days. Early stop of crystalloid + pressor escalation is a better strategy than indefinite bolusing. (See ANDROMEDA-SHOCK 2019.)
Each liter of NS has Cl⁻ 154 — 50% higher than plasma. Large volumes → strong-ion-gap acidosis (non-AG) that is distinct from sepsis-driven lactic acidosis. This is why balanced crystalloids are now preferred for high-volume resuscitation. The acidosis isn't benign: it impairs splanchnic perfusion and coagulation.
Calcium in LR (3 mEq/L) can precipitate with ceftriaxone (fatal pulmonary emboli in neonates reported) or cause clot formation with citrated blood products. Use a separate line, or switch to Plasma-Lyte (no Ca) if you only have one line.
D5 NS at 125 mL/h delivers ~150 g glucose/day — enough to raise glucose 60–100 mg/dL in an insulin-resistant patient. On TPN or high dextrose maintenance, BG rises are common and blamed on "stress hyperglycemia" when they're iatrogenic. Choose the non-dextrose option if glycemic control matters.
Press 1–6 to jump between stages. Press / to focus the simulator.