Ambient temperature and ambulatory blood pressure: An hourly-level, longitudinal panel study

Abstract

Background: Variations of blood pressure (BP) related to air temperature have been reported previously; however, no evidence is available regarding the association of hourly ambient temperature with ambulatory blood pressure.

Methods: We conducted a longitudinal panel study among 1895 patients from an outpatient department who received repeated ambulatory blood pressure monitoring in Urumqi, China between July 2020 and December 2021. We obtained hourly ambient temperature from the nearest monitoring station to the residential address, and measured 4 ambulatory blood pressure indicators. Linear mixed-effect model combined with distributed lag models were applied to investigate the cumulative associations of hourly temperature with BP.

Results: A total of 97,466 valid blood pressure measurements were evaluated. We observed almost linear and monotonically decreasing relationships between temperature and blood pressure. The effects occurred in the same hour, attenuated thereafter and became insignificant approximately 36 h. A 10 °C decrease in temperature was significantly associated with increments of 0.84 mmHg in systolic blood pressure, 0.56 mmHg in diastolic blood pressure, 1.38 mmHg in mean arterial pressure, and 0.66 mmHg in pulse pressure over lag 0 to 36 h. Stronger associations were found among patients of female sex, age between 18 and 65 years, overweight or obesity, minority, less education or in the cold season, as well as those without hypertension or with coronary heart disease, or did not take anti-hypertension medication.

Conclusion: Our study provides robust evidence that hourly ambient temperature is inversely associated with ambulatory blood pressure. It also highlights a linear relationship between decreased ambient temperature and elevated BP, which may have implications for the prevention and management of hypertension in susceptible populations.

Keywords: Ambient temperature; Ambulatory blood pressure; Distributed lag model; Linear mixed–effect model.