Table 3 Sensitivity analyses for the association of CRAE and AVR with leukocyte telomere length, stratified by sex and BMI.

From: Retinal microcirculation and leukocyte telomere length in the general population

Outcome

BMIc

Men

Women

P-Interaction sex × BMI × LTL

n

Estimate (95% CI)

n

Estimate (95% CI)

CRAE a

Excluding current and former smokers

23.6

58

−1.40 (−4.65, 1.86)

62

0.04 (−4.12, 4.20)

0.033

 

29.4

58

−7.10 (−11.16, −3.03)

62

1.02 (−3.56, 5.61)

 

Excluding hypertension

23.1

56

−0.90 (−4.96, 3.15)

87

−1.23 (−4.99, 2.54)

0.0098

 

28.5

56

−8.80 (−13.50, −4.10)

87

−0.87 (−4.20, 2.45)

 

Excluding diabetes

23.8

119

−1.50 (−4.17, 1.17)

123

−0.87 (−3.79, 2.05)

0.006

 

29.3

119

−5.80 (−8.59, −3.02)

123

0.36 (−2.26, 2.97)

 

AVR b

Excluding current and former smokers

23.6

58

−0.004 (−0.018, 0.010)

62

−0.006 (−0.024, 0.011)

0.082

 

29.4

58

−0.015 (−0.032, 0.002)

62

0.006 (−0.014, 0.026)

 

Excluding hypertension

23.1

56

−0.002 (−0.021, 0.018)

87

−0.006 (−0.023, 0.012)

0.036

 

28.5

56

−0.029 (−0.052, −0.007)*

87

−0.002 (−0.018, 0.014)

 

Excluding diabetes

23.8

119

−0.006 (−0.018, 0.007)

123

−0.008 (−0.022, 0.005)

0.0074

 

29.3

119

−0.020 (−0.033, −0.007)

123

0.003 (−0.009, 0.015)

 
  1. Abbreviations: BMI, body mass index; CI, confidence interval; LTL, leukocyte telomere length; CRAE, central retinal arteriolar equivalent; AVR, arteriole-to-venule ratio. Models adjusted for age, sex, systolic blood pressure, smoking status, BMI as fixed effects and adjusted for participants nested within a family cluster as random effect. aEstimates presented as difference in arteriolar diameter (μm) for a 20% shorter LTL; bEstimates presented as difference in arteriole–to–venule ratio for a 20% shorter LTL; cBMI in kg/m2 provided for the 25th and 75th percentile. Significance for estimates: *P ≤ 0.05, P ≤ 0.01, P < 0.0001.