Thiazides and magnesium

A 24-hour episode of profuse, watery diarrhea brought her to the Emergency Department gravely ill. On admission she was hypotensive and tachycardic, but had no localizing abdominal symptoms. Her admitting lab studies were sodium 136, potassium 3.5, chloride 98, bicarbonate 16, creatinine 3.1, and estimated glomerular filtration rate 16. Five weeks earlier, the corresponding lab values at a routine office visit were 139, 3.3. 95, 30, 0.7 and 88. Rehydration was commenced immediately and she responded well.

On admission her serum calcium (uncorrected) was 6.5 mg/dL (9.3, five weeks earlier), phosphate 5.4, and her magnesium 0.4 mg/dL (reference interval 1.7-2.5).

With those values she was fortunate that she did not have a lower bicarbonate level because rapid correction intravenously would almost certainly have resulted in tetany. Low bicarbonate is protective against the muscular effects of hypocalcemia.

Six weeks later her renal function has been restored to baseline values, electrolytes are normal as are calcium and inorganic phosphate. The residual problem was difficulty in maintaining normal serum magnesium levels.

There is no good laboratory assessment of body stores of magnesium and these may be depleted while the serum level remains normal for quite some time. Hypomagnesemia is a clear indication that things are not OK. Oral magnesium is OK long term but IV magnesium is needed initially and may be required for several days.

Would even a 24-hour episode of watery diarrhea sufficient to cause such marked dehydration be sufficient to cause such low values for calcium and magnesium? Probably not, so another cause needs to be sought. The culprit was almost certainly the hydrochlorothiazide she had been taking for several years for management of mild hypertension. We are all familiar with the potential adverse effects of thiazides on sodium and potassium but most often neglect the effect on magnesium. Patients presenting with cramps or arrhythmia while on thiazides need to be evaluated for low magnesium, particularly if the potassium is normal. The missed clue (and hindsight is always 20/20 vision) was the serum potassium of 3.3 five weeks prior to admission. Not of itself a worryingly low number but important in retrospect.

Why was the calcium so low on presentation, given the degree of dehydration? Hypomagnesemia inhibits parathyroid hormone secretion so she was physiologically profoundly hypoparathyroid. This is an uncommon complication of thiazide therapy. Far more often I see this in patients referred for osteoporosis and a low 25-hydroxyvitamin D level. Such low levels should promote increased PTH secretion and if the PTH is not elevated serum magnesium should be measured. Remember that serum levels of magnesium are not very reflective of body stores of magnesium so that a combination of a low 25-hydroxyvitamin D and a normal PTH should prompt you to offer a magnesium supplement no matter what the serum level. Don’t forget the vitamin D supplement as well.