Confusion, Seizure, and a Sodium of 110 after Starting a Thiazide
A 67-year-old woman with hypertension is brought to the emergency department after her daughter found her confused and vomiting. Hydrochlorothiazide 25 mg daily was started three weeks ago; she takes no other medication and denies alcohol use. She is disoriented to place and time, afebrile, pulse 96/min, blood pressure 108/64 mmHg, weight 52 kg. Mucous membranes are dry, there is no jugular venous distension and no edema. During assessment she has a 40-second generalized tonic-clonic seizure that stops spontaneously; she is postictal afterwards. Point-of-care glucose is 96 mg/dL. Serum sodium 110 mmol/L, potassium 3.1 mmol/L, chloride 78 mmol/L, bicarbonate 26 mmol/L, BUN 14 mg/dL, creatinine 0.9 mg/dL, serum osmolality 232 mOsm/kg. Urine sodium 62 mmol/L, urine potassium 28 mmol/L, urine osmolality 540 mOsm/kg. Non-contrast CT of the head is unremarkable, and she is not on any other medication known to cause hyponatremia.
1. What is the mechanism of this patient's hyponatremia, and how do the urine studies change your assessment of her volume status?
Read the worked answer
The serum osmolality of 232 mOsm/kg confirms true hypotonic hyponatremia (calculated osmolality ≈ 2×110 + 96/18 + 14/2.8 ≈ 230), so this is not pseudohyponatremia or hyperglycemic dilution.
Mechanism: thiazides produce hyponatremia through two additive effects - blockade of the distal convoluted tubule sodium-chloride cotransporter impairs maximal free-water clearance, and the resulting mild volume contraction plus nausea/vomiting is a potent non-osmotic stimulus for antidiuretic hormone. A urine osmolality of 540 mOsm/kg shows ADH is acting and the kidney cannot excrete the water load.
Interpretation of the urine sodium: a value of 62 mmol/L does not exclude volume depletion here. Thiazides act downstream of the dilution segment and cause obligate natriuresis, so urine sodium is unhelpful for volume assessment in a patient on the drug. The examination (dry mucosae, no edema, no JVD) suggests mild contraction, but she is not shocked.
Before accepting 'thiazide-induced' as complete, send TSH and a morning cortisol to exclude hypothyroidism and adrenal insufficiency, both of which cause hypoosmolar, ADH-driven hyponatremia with high urine osmolality. Her potassium of 3.1 mmol/L is also relevant: hypokalemia both predisposes to osmotic demyelination and shifts sodium upward as it is corrected.
2. The seizure has just stopped. Outline your management for the next hour, including a calculation of how much 3% saline to give and the sodium target you are aiming for.
Read the worked answer
This is symptomatic hyponatremia with a seizure - an indication for hypertonic saline, not fluid restriction and not a vasopressin receptor antagonist.
Airway and monitoring. Protect the airway, apply oxygen, continuous cardiac and pulse oximetry, hourly (initially every 1–2 hours) serum sodium, and move to a resuscitation area.
Hypertonic saline. Give 3% sodium chloride 100–150 mL as an intravenous bolus over 10–20 minutes. Reasoning with the Adrogué–Madias relationship: total body water ≈ 0.5 × 52 kg = 26 L; the change in sodium per litre of 3% saline (513 mmol/L) ≈ (513 − 110)/(26 + 1) ≈ 15 mmol/L. A 150 mL bolus therefore raises sodium by only about 2.2 mmol/L - enough to shrink brain cells and stop seizures, but not enough to cause harm. Repeat the bolus up to two or three times until the seizure stops or the sodium has risen by 4–6 mmol/L.
Targets. Aim for a 4–6 mmol/L rise within the first 6 hours, a maximum of 8 mmol/L in 24 hours, and never more than 10 mmol/L in 24 hours or 18 mmol/L in 48 hours.
Do not: rapid correction beyond these limits, hypertonic saline continued after symptoms resolve, or hypotonic fluids during the acute phase. Correct potassium cautiously and anticipate that potassium replacement itself will raise the serum sodium.
3. Six hours later the repeat sodium is 124 mmol/L; she is now alert, oriented and seizure-free. What has gone wrong, what is the immediate risk, and what do you do now?
Read the worked answer
The sodium has risen 14 mmol/L in 6 hours - well beyond the 8 mmol/L/24-hour ceiling and far faster than the 4–6 mmol/L initially targeted. This is overcorrection and she is at real risk of osmotic demyelination. Her risk factors are additive: sodium below 105 mmol/L, female sex, hypokalemia at presentation, vomiting with poor intake, and a thiazide (a recognised precipitant in this syndrome).
Immediate steps:
- Stop all hypertonic saline and stop any potassium correction that is driving the sodium up.
- Actively re-lower the sodium: give 5% dextrose in water (typically 10 mL/kg over 1 hour, or roughly 3–5 mL/kg/h) and, if the rise continues, add desmopressin 1–2 micrograms subcutaneously or intravenously to lock the kidneys and allow controlled re-lowering.
- Target a fall of about 1 mmol/L per hour until the net rise over the first 24 hours is under 8–10 mmol/L; the exact goal is a re-lowered sodium that keeps her within the safe 24-hour window.
- Recheck sodium hourly, watch for dysarthria, dysphagia, ataxia, behavioural change or quadriparesis, and obtain MRI if any neurological deterioration occurs.
What not to do: simply watch and repeat the sodium in 6 hours. She is currently asymptomatic, which is precisely when re-lowering prevents locked-in syndrome. Document the calculation, the desmopressin decision, and the reasoning in the record.