Increased Risk of Drug-Induced Hyponatremia during High Temperatures

Anna K. Jönsson, Henrik Lövborg, Wolfgang Lohr, Bertil Ekman, Joacim Rocklöv

To investigate the relationship between outdoor temperature in Sweden and the reporting of drug-induced hyponatremia to the Medical Products Agency (MPA). All individual adverse drug reactions (ADR) reported to MPA from 1 January 2010 to 31 October 2013 of suspected drug-induced hyponatremia and random controls were identified. Reports where the ADR had been assessed as having at least a possible relation to the suspected drug were included. Information on administered drugs, onset date, causality assessment, sodium levels, and the geographical origin of the reports was extracted. A case-crossover design was used to ascertain the association between heat exposure and drug-induced hyponatremia at the individual level, while linear regression was used to study its relationship to sodium concentration in blood. Temperature exposure data were obtained from the nearest observation station to the reported cases. During the study period, 280 reports of hyponatremia were identified. More cases of drug-induced hyponatremia were reported in the warmer season, with a peak in June, while other ADRs showed an opposite annual pattern. The distributed lag non-linear model indicated an increasing odds ratio (OR) with increasing temperature in the warm season with a highest odds ratio, with delays of 1-5 days after heat exposure. A cumulative OR for a lag time of 1 to 3 days was estimated at 2.21 at an average daily temperature of 20 °C. The change in sodium per 1 °C increase in temperature was estimated to be -0.37 mmol/L (95% CI: -0.02, -0.72). Warm weather appears to increase the risk of drug-induced hyponatremia.