Abstract
This study was performed to compare the effect of one month's treatment with hydrochlorothiazide (25 mg) in combination with either amiloride (2.5 mg) or triamterene (50 mg) on plasma sodium levels in elderly people in institutional care. Fifty residents of NHS nursing or social service residential care established on diuretics for congestive cardiac failure and aged 64 years or over were recruited. Forty-one patients were included in the final data analysis. Patients on hydrochlorothiazide/amiloride had a significantly lower plasma sodium (137 vs 139 mmol/l, 95% confidence interval for difference between medians 0-2 mmol/l) than those on hydrochlorthiazide/triamterene (P = 0.01). In equivalent potassium-retaining doses, amiloride is associated with significantly lower plasma sodium levels than triamterene, when given in combination with hydrochlorothiazide in elderly patients with congestive cardiac failure. This finding adds weight to uncontrolled observations implicating thiazide/amiloride diuretic combinations in causing serious hyponatraemia. This danger, although uncommon, should perhaps influence prescribing habits in an at-risk population.