Introduction
Cardiac arrhythmias are a heterogeneous group of cardiac conditions that frequently require intensive care treatment and place a substantial burden on hospital resources. Although traditional risk factors for acute arrhythmic events are well established, the role of short-term atmospheric exposures as potential triggers remains incompletely understood at the population level. Atmospheric conditions such as heat stress, cold spells, pressure changes, and air pollutants may acutely destabilize cardiac electrophysiology and precipitate arrhythmic episodes.
Methods
Within the ALERT-ITS project, we performed a retrospective population-based time-series analysis based on anonymized routine data from the German Allgemeine Ortskrankenkasse (AOK, 2009–2023). Daily intensive care unit (ICU) admissions for cardiac arrhythmias were identified from inpatient records using predefined ICD-10-GM criteria (ICD-10-GM: I47.0, I47.1, I47.2, I47.9, I48.00, I48.01, I48.10, I48.11) and stratified by sex and age (≥60 vs. <60 years). Cases were assigned to nationwide biometeorological regions, with analyses focusing on Southeast Germany. Meteorological data were obtained from ERA5 reanalysis and air-quality data from the Copernicus Atmosphere Monitoring Service, covering 31 atmospheric parameters. Associations with daily case counts were modelled using generalized additive models with quasi-Poisson distribution, incorporating distributed lag structures (meteorological variables up to 10 days, pollutants up to 7 days) and adjusting for seasonality and long-term trends.
Results
During the study period (2009–2023), a total of 49.284 ICU admissions for arrhythmias were identified in Southeast Germany (48,9% female, 74,8% aged ≥60 years), corresponding to a mean annual rate of 70 per 100,000 insured individuals per year. From the pool of 31 atmospheric variables, significant associations were identified for wind days (WD), tropical nights (TN), cold wave duration (CWD), heat day (HD), heat wave duration (HWD), ozone (O₃), daily mean sea level pressure (SLP), and wind chill index (WCI). While SLP was significantly associated in all subgroups (men ≥60 years (p=0.014), women ≥60 years (p=0.029), men <60 years (p=0.006) and women <60 years (p=0.003)), no further significant predictor was identified in male subgroups. However, women ≥60 years showed significant associations with HD (p=0.015), O₃ (p=0.001) and CWD (p=0.016). In women <60 years, significant associations were found for HD (p=0.049), TN (p=0.008), HWD (p=0.005) and WCI (p=0.043). The strongest association was observed for women ≥60 years in O₃, with a relative risk of 0.967 (95% CI 0.947–0.987), corresponding to a 3.3% reduction on days with an increase of 10 ppb.
Discussion
This analysis identifies environmental factors associated with short-term fluctuations in ICU admissions for arrhythmias in Southeast Germany, with distinct patterns across sex and age subgroups. While SLP was significantly associated with admissions in all subgroups, heat stress (HD, TN, HWD) were significantly associated exclusively in female subgroups. Heat-related exposures may promote arrhythmogenic conditions through electrolyte shifts and increased sympathetic tone, while cold spells and pressure changes may act via vagal modulation and hemodynamic stress. These subgroup-specific patterns underline the need for targeted risk stratification and integrating environmental forecasting into clinical preparedness strategies.
