Effects of short-term exposures to air temperature and temperature variability on blood pressure and heart rate in the German National Cohort
I Moreno Velasquez, N Nikolaou, K Wolf, A Peters, S Breitner, A Schneider, T Pischon, N A K O On Behalf Of The Nako StudyAbstract
Background
Temperature variability (TV) is an emerging cardiovascular risk factor, yet the independent effects of air temperature (aT) and TV on hemodynamic parameters in European populations remain unclear.
Aim
To investigate the association of short-term aT and diurnal TV individually and jointly with systolic (SBP) and diastolic blood pressure (DBP), pulse pressure (PP), mean arterial pressure (MAP) and heart rate (HR) among participants of the German National Cohort (NAKO).
Methods
Baseline BP and HR measurements from 156,353 NAKO participants (2014-2019) were included. Daily mean, minimum and maximum aT were estimated via a multistage regression-based 1 × 1 km spatiotemporal model and linked to residential addresses. TV was defined as the diurnal range (maximum–minimum aT) over 0-6 days. Associations were assessed using generalized additive models, expressed as % differences (95%-CI) in outcome means per 1 °C decrease in mean aT at immediate (lag 0), delayed (lags 1 to 6), and cumulative (lag 0-6) exposures. Distributed lag-nonlinear models were used to estimate cumulative differences (95%-CI) in outcomes associated with diurnal TV, relative to the minimum predicted effect temperature (MET). We evaluated effect modification by age, sex, and season. Stratified analyses were conducted by antihypertensive medication use, cardiovascular disease (CVD), diabetes, and BP categories. Diurnal TV effects were further stratified by aT quartiles. Mutually adjusted models assessed the independent effects of aT and diurnal TV.
Results
Lower aT was associated with higher BP across all lags, with strongest effects at cumulative lag 0-6: per 1°C aT decrease, SBP increased by 0.13% (95%-CI 0.11;0.14), DBP by 0.10% (0.08;0.12), PP by 0.17% (0.14; 0.20), and MAP by 0.11% (0.10;0.13). We observed stronger associations for females, participants ≥60 years, with CVD or examined during warm season. Small-to-moderate TV increases were associated with higher SBP, DBP and MAP, with stronger effects in males and warmer aT quartiles. After mutual adjustments, both aT and TV were independently associated with BP and HR. While the association between aT and BP outcomes remains inverse after adjusting for TV, higher TV was associated with progressively higher BP only after accounting for aT, with the largest effects observed for SBP. Over cumulative lag 0-6, TV≈ 4°C and 15°C, compared to MET, was associated with a 0.70% (95%-CI 0.15;1.26) and 1.03% (95%-CI 0.31;1.74) higher SBP, respectively. Relative to MET, aT was associated with heat-related increases in HR. Both low and high TV were associated with higher HR; this association remained unchanged after mutual adjustments.
Conclusion
Both lower mean aT and higher diurnal TV were independently associated with higher BP. Our results further suggest that increasing diurnal TV is related to higher BP and HR, independent of aT, and that these associations appeared stronger at higher aT.