Differences in impact of current and former shift work on cardiovascular risk factors, carotid atherosclerosis, and...

The association of shift work (SW) and disrupted circadian rhythm with markers of large artery atherosclerosis and cerebral small vessel disease is uncertain. We aimed to study the separate association of current and former SW with these markers.

We included participants from the population-based Hamburg City Health Study. SW was defined by monthly working hours between 06:00 pm and 07:00 am containing night shifts for at least 12 months. Cross-sectional data were obtained from structured questionnaires, laboratory analyses, physical examinations, brain magnetic resonance imaging, and carotid ultrasound. We performed multivariable regression analysis with carotid intima-media thickness (CIMT), and peak-width skeletonized mean diffusivity (PSMD) as dependent variables.

Three hundred and forty-four current, 238 former, and 7162 never-shift workers were included. The median age was 60 years for both current and former shift workers, and total duration of SW was comparable for the two groups. Current shift workers were less frequently female (27.3% vs. 44.5%; p < .001), had more frequent hyperlipidemia (31.5% vs. 22.3%; p = .024), and diabetes (16.2% vs. 3.2%; p < .001). After adjustment for age and sex, reduced quality of sleep (β = 1.61, p = .001) and low education (β = 2.63, p < .001) were associated with current but not former SW. Adjusted for age and sex, the current SW was associated with higher CIMT (β = 0.02, p = .001) and PSMD (β = 9.06e-06, p = .006), whereas former SW was not. Adjusted for risk factors, current SW remained associated with PSMD (β = 9.91e-06, p = .006) but not with CIMT.

Current SW was associated with CIMT and with PSMD, with the latter association remaining after adjustment for risk factors. Former SW showed no associations with CIMT or PSMD. This may indicate that current SW is linked with increased neurovascular risk through disrupted circadian rhythms.

The trial was submitted at http://www.clinicaltrials.gov, under NCT03934957 on January 4, 2019. The first participant was enrolled in February 2016.