Back
Short sleep duration is associated with declined estimated glomerular filtration rate (eGFR) in a large cohort of women; however, the association of sleep duration with eGFR in a nationally representative population is unknown. We aim to examine the association between sleep duration and eGFR in the general population.
A cross-sectional study using 2017-2020 National Health and Nutrition Examination Survey (NHANES) included self-reported sleep duration on weekdays and weekends and eGFR. The association between the quartile (Q) of sleep duration on both weekdays and weekends with eGFR was examined by multiple linear regression.
Of 9,693 participants ≥18 years old, mean±SD age was 50±19 years and 51% were female. Mean sleep duration on the weekdays and weekends were 7.62±1.69 and 8.30±1.81, respectively (Mean difference -0.68; 95%CI -0.71, -0.64; P <0.001). Mean sleep duration on the weekdays in Q1 to Q4 were 6.115571±1.070067, 7.5±0, 8.14±0.22, and 9.76±1.03 hours, respectively, and mean corresponding weekend sleep duration were 6.16±1.06, 7.89±0.21, 8.96±0.30, and 10.81±1.023 hours, respectively. Mean eGFR was 101.81±38.69 mL/min/1.73 m2. Median eGFR (IQR) was 98.43 mL/min/1.73 m2 (76.16, 122.73). Among 9,421 participants with reported diabetes status, 1,423 and 7,998 had and did not have diabetes, respectively (Figure 1). Compared to participants with sleep duration in Q1, eGFR were 3.3 to 10.7 mL/min/1.73 m2 significantly higher in weekend sleep duration in Q2 to Q4, but not in weekday sleep duration (Figure 2). After adjusting for age, gender, race, SBP, DBP, pulse, BMI, urinary albumin:creatinine ratio, hemoglobin, ferritin, hemoglobin A1c, and the interaction term between quartile of sleep duration (on both weekdays and weekends) and race (white vs. non-white), the direction of the weekend sleep duration – eGFR association remained but different in the magnitude with 5.1 to 4.8 mL/min/1.73 m2 significantly higher eGFR in weekend sleep duration in Q2 to Q4. There was no statistically significant association between weekday sleep duration and eGFR. In addition, race was identified as an effect modifier with a decreased weekend sleep duration observed in White with sleep duration in Q2 to Q4 (βweekends_Q2, Q3, and Q4 -5.9, -6.1, and -5.5; Pinteraction 0.007, 0.005, and 0.031). There was no indication for effect modification of the weekday sleep duration with other covariates.
Sleep duration on the weekends, but not on the weekdays, is positively associated with eGFR in a general representative population. However, the weekend sleep duration and eGFR is inversely related in White. Several factors including environments on the weekends and races may contribute to sleep quality and subsequently eGFR. Further longitudinal studies are required to examine the potential involved mechanism of sleep duration at different times of the week and eGFR.