Up-to-date, there have been 7 case reports of hypomagnesaemia associated with hypocalcaemia and hypokalaemia in patients on long-term treatment with proton-pump inhibitors (PPIs). The suggestion is that these reports present a tip of an iceberg in clinical practice. Inspired by a case of severe symptomatic electrolyte loss, which was corrected by discontinuation of PPIs, we performed an audit of a potential relationship between hypomagnesaemia and PPIs in our hospital.
From October 2008 until October 2009, our biochemistry department reported 34 cases of low magnesium (Mg) in the outpatient department and on the medical admission unit (26 female, age 65±14). Therapeutic information was available in 28/34 and included PPIs in 22/28. Other potential contributors to electrolyte disturbance could be speculated in 11/30 for whom the full medical history was available. Mean serum Mg was 0.5±0.1 mmol/l. Following treatment, which included PPI withdrawal in 1/22 and Mg infusion in 15/16 patients, it rose to 0.7±0.4 mmol/l, remaining subnormal in 10/18 patients in whom serum Mg was repeated. Serum calcium (Ca) was measured in 32/34 patients and was 2.08±0.24 mmol/l (presented as corrected Ca), with readings below reference range in 13/32. It increased to 2.24±0.13 mmol/l in 28/34 patients in whom the measurement was repeated, remaining low in 5/28 despite Ca and vitamin D supplements. Hypomagnesaemia was associated with hypokalaemia in 1/32 patients (mean 3.8±0.7 mmol/l). Three out of 18 patients were symptomatic with tetany (2) and seizures (1).
There is a little awareness of the relationship between PPI treatment and electrolyte disturbances in clinical practice. Despite hypomagnesaemia being a fairly common finding, PPI therapy rarely figures in the differential diagnosis. Without recognition of possible causality between the two, it is difficult to achieve electrolyte normalisation purely with replacement therapy. Greater recognition of this phenomenon is required and H2 antagonist substitution should be considered.