BACKGROUND: Over the past 50 years, heatwaves have become more common and severe. Some medications may sensitize fragile patients, such as those with heart failure (HF), to the effects of extreme heat events.
OBJECTIVES: To determine if medications common in HF patients may increase the short-term risk of heat-related adverse events during the warm summer months in the US.
METHODS: Linking US Medicare data with daily maximum surface air temperature measures (tasmax) based on residence ZIP code, we identified patients at their first hospital discharge with a primary or secondary HF diagnosis during June–August of years 2007–2012. We followed individuals from discharge to August 31 and characterized their periods of use of medication classes of interest—angiotensin converting enzyme inhibitors (ACE), angiotensin receptor blockers (ARB), loop diuretics (LD), antipsychotics (AP), beta blockers (BB), and anticholinergic agents (AC). We defined heatwaves occurring during follow-up as two consecutive days with tasmax above the 95th percentile in the distribution of historical data. We recorded heat-related hospitalizations (heat exhaustion, heat fatigue, heatstroke, dehydration, hyperosmolality, and excessive exertion), recurrent HF hospitalization, and mortality during follow-up. We performed self-controlled case series analysis to estimate risk ratios (RR) and 95% confidence intervals (95% CI) of the association of current medication use with heat-related adverse events, accounting for heatwaves occurring follow-up.
RESULTS: We identified 146,642 discharged HF patients (64.5% female; 78.4% white; mean age = 81.2, SD 8.3). 33% experienced a heatwave during follow-up. Medication exposure varied widely: LD, 75%; AC, 75%; BB, 71%; ACE/ARB, 61%; AP, 9%. A heat-related hospitalization occurred in 4.1% of patients. After accounting for heatwaves, all medication classes were associated with moderately increased heat-related hospitalizations—RR ranged from: AP RR = 1.43 (1.17–1.74) to ACE/ARB RR = 1.77 (1.59–1.96). AC and AP medications were associated with subsequent HF hospitalizations—AC RR = 1.18 (1.11–1.26), AP RR = 1.18 (1.04–1.34).
CONCLUSIONS: The studied medication classes were associated with increased hospitalizations for heat-related illness among HF patients during the summer months. A larger scale study incorporating other climate and clinical factors is needed to confirm the results in this and other clinical populations.
