hypernatraemia and diabetes insipidus

in review 5 min read Updated 2026-10-27
hypernatraemia and diabetes insipidus

Hypernatraemia = water deficit relative to sodium, almost always from no access to water (intubated, delirious, frail) plus losses. Lower chronic hypernatraemia by ~10 mmol/L/day, mostly with enteral water or D5W; adults are more often undertreated than overcorrected. Hypotonic polyuria (>3 L/day or >40–50 mL/kg/day, urine <800 mOsm/kg) → AVP deficiency (central DI) vs AVP resistance (nephrogenic DI) vs primary polydipsia. Stimulated copeptin has replaced the water-deprivation test where available.

do not miss
  • Acute salt poisoning (minutes–hours: salt ingestion, hypertonic saline, dialysis error) → rapid D5W + emergency haemodialysis; risk is intracranial haemorrhage, not overcorrection
  • Post-pituitary surgery DI in a patient who can’t drink → hypernatraemia within hours
  • Undertreatment: cerebral oedema from fast correction is reported only in children; adults are usually undertreated

causes

mechanismexamplesurine osm (mOsm/kg)
renal water lossAVP-D, AVP-R, osmotic diuresis (glucose, urea, mannitol, post-obstruction)<300 (DI) or 300–600 (osmotic)
extrarenal water lossfever, burns, ventilation, diarrhoeahigh (>600–700)
poor intakeimpaired thirst or access (elderly, ICU, hypothalamic lesion)high
sodium gainhypertonic saline, NaHCO₃, salt ingestionvariable

correction

water deficit (L)=TBW×(Na140−1)\text{water deficit (L)} = \text{TBW} \times \left(\frac{\text{Na}}{140} - 1\right)

TBW ≈ 0.6 × weight (young men), 0.5 (women, older men), 0.45 (older women). Add ongoing renal and insensible losses. Rule of thumb: 3 mL/kg electrolyte-free water lowers Na⁺ ~1 mmol/L (no ongoing urine water loss). Monitor Na⁺ rather than trusting the formula.

onsettypical causetarget
minutes–hourssalt poisoningrapid D5W + emergency HD; overcorrection not known to harm
1–2 daysunreplaced urine water (DI, glycosuria)lower ~2 mmol/L/h to 145; replace or stop ongoing losses
≥2 days or unknownhypodipsia, impaired mental statuslower ~10 mmol/L/day; replace ongoing losses
  • The ≤0.5 mmol/L/h limit comes from rehydration seizures in children; no adult limit is established
  • Enteral free water if the gut works; otherwise D5W (watch glucose)
  • Hypovolaemic and shocked: isotonic fluid first, then water
  • Na⁺ every 4–6 h while correcting

polyuria and DI

2022 nomenclature: arginine vasopressin deficiency (AVP-D) = central DI; arginine vasopressin resistance (AVP-R) = nephrogenic DI.

AVP-D (central)AVP-R (nephrogenic)
causespituitary/hypothalamic surgery, trauma, craniopharyngioma, infiltrative (sarcoid, LCH), idiopathiclithium, hypercalcaemia, hypokalaemia, tubulointerstitial disease, post-obstruction, congenital (AVPR2, AQP2)
response to desmopressinurine osm rises >50%, volume fallslittle or none
treatmentdesmopressin (µg doses: e.g. intranasal 10–20 µg, oral 100–400 µg/day) treat the cause (correct Ca²⁺, K⁺); low-solute (low Na, low protein) diet; thiazide; amiloride for lithium
  • Confirm hypotonic polyuria on a 24-h collection. Urine osm >800 → solute diuresis (glucose incl. SGLT2i, urea from protein/catabolism/steroids, mannitol, diuretics), not DI
  • Baseline bloods: Na⁺ >147 / osm ≥300 → AVP-D or AVP-R; Na⁺ <135 / osm ≤280 → primary polydipsia (desmopressin dangerous here → hyponatraemia). Most patients are in between and need testing
  • Already hypernatraemic (Na⁺ >147) with hypotonic polyuria → the osmotic stimulus is already there, so no deprivation or stimulation test: copeptin ≤4.9 pmol/L = AVP-D, >21.4 = AVP-R. Without copeptin, the desmopressin response (urine osm rise >50% = AVP-D) is the usual substitute
testresultinterpretation
basal copeptin (unstressed)>21.4 pmol/LAVP-R
hypertonic saline-stimulated copeptin (3% 250 mL over 15 min, then weight-adapted infusion; Na⁺ every 30 min; sample once >149)≤4.9 pmol/LAVP-D (accuracy 96–97%)
>4.9 pmol/Lprimary polydipsia
arginine-stimulated copeptin (0.5 g/kg in 500 mL NS over 30 min)≤3.0 / >5.2 pmol/LAVP-D / primary polydipsia (in between → indeterminate)
water deprivation (≥16 h) then desmopressinurine <300 → rise >50% / <50%complete AVP-D / AVP-R
urine 300–800 → rise ≥9% / <9%partial AVP-D / partial AVP-R or primary polydipsia
  • Water deprivation accuracy only ~77%: often fails to produce enough osmotic stimulus. Hypertonic saline needs rapid Na⁺ turnaround and continuous monitoring; arginine is shorter (1 h) and better tolerated. Canadian copeptin availability is limited

  • Desmopressin hyponatraemia: ~30% of outpatients with AVP-D on routine bloods. Start with a night-time dose; check Na⁺ in the first days; counsel against excess drinking. Desmopressin escape: delay or omit a dose weekly to allow an aquaresis

  • AVP-D patients drinking to thirst rarely get hypernatraemic. Risk is when thirst or access is lost: surgery, reduced consciousness, adipsic AVP-D (hypothalamic osmoreceptor damage)

post-pituitary surgery: triphasic response
DI (days 1–5) → SIADH from release of stored AVP (days 5–10) → permanent DI if most AVP neurons are lost. Desmopressin given in phase 1 is dangerous in phase 2 → daily Na⁺.
desmopressin and lithium
  • Desmopressin 2–4 µg IV/SC
  • Lithium AVP-R: amiloride blocks lithium entry via ENaC in the collecting duct
  • Thiazides reduce polyuria in AVP-R through mild volume contraction and more proximal reabsorption (paradoxical)

Prepared 2026-10-27. Not yet verified — check every number against the sources.

Key references

Suggest a correction hypernatraemia and diabetes insipidus