Hyponatraemia in very old nonhospitalised people: association with drug use.

OBJECTIVE

Hyponatraemia is one of the major problems in geriatric inpatients. However, in nonhospitalised elderly, the preponderance of hyponatraemia and the importance of the effect of drug intake on serum sodium concentrations are little known. This study investigated the prevalence of hyponatraemia in very old nonhospitalised people, controlling for factors that may induce hyponatraemia (especially drug use).

METHODS

Data on serum sodium concentration, health and drug use were retrieved for 185 persons aged 92 to 93 years (the 1905 cohort) and 147 persons aged 100 years (the centenarian cohort) participating in two major population-based studies of elderly people in Denmark. Data were analysed by comparing median serum sodium concentrations between users and nonusers of various drugs after controlling for the influence of age, sex, cancer, heart failure, hypothyroidism, renal failure and smoking. Furthermore, the preponderance of drug use in the patients with clinically relevant hyponatraemia was compared with that in persons with normal serum sodium concentrations.

RESULTS

Median serum sodium concentration was 140 mmol/L for the centenarians and 141 mmol/L for the 1905 cohort. In total, 19 persons had hyponatraemia (serum sodium concentration < or =134 mmol/L). There was no association between median serum sodium concentration and any of the above-mentioned diseases, or sex or smoking. Of the drugs generally known to cause hyponatraemia, only omeprazole and oral antidiabetic agents were associated with significantly lower median serum sodium concentrations (difference 3 mmol/L). Use of thiazide diuretics was significantly more common than expected in persons with hyponatraemia compared with persons with a normal serum sodium concentration (7 of 19 vs 46 of 270 individuals). Furthermore, the results suggested that digoxin and lactulose might be associated with a lowered median serum sodium concentration.

CONCLUSION

This study demonstrates that severe hyponatraemia was rarely seen in a population-based sample of very old persons and that drugs have only a limited influence on serum sodium concentration. The only drug class associated with clinically relevant hyponatraemia was thiazide diuretics, which were used by significantly more persons with hyponatraemia. Furthermore, this study suggests that digoxin and lactulose use is associated with lower serum sodium concentrations in the elderly.