Skip to main content
. 2019 Nov 18;9:16972. doi: 10.1038/s41598-019-53546-y

Table 3.

Association of longitudinal risk profile trajectory clusters with different adipose tissue traits and impact of additional adjustment for risk factors obtained at Exam 1 or Exam 3.

outcome Model Cluster I Cluster II Cluster III
estimate 95%-CI p-value estimate 95%-CI p-value
TAT fully adjusted Ref. 3.90 [2.89, 4.91] <0.01 9.49 [8.20, 10.78] <0.01 R2 = 0.54
fully adjusted + risk profile Exam 1 Ref. 3.21 [2.14, 4.29] <0.01 6.25 [4.73, 7.77] <0.01 R2 = 0.65
fully adjusted+risk profile Exam 3 Ref. 0.39 [−0.26, 1.04] n.s 0.96 [0.09, 1.82] 0.03 R2 = 0.89
VAT fully adjusted Ref. 1.30 [0.84, 1.75] <0.01 3.32 [2.74, 3.90] <0.01 R2 = 0.63
fully adjusted + risk profile Exam 1 Ref. 1.16 [0.63, 1.69] <0.01 2.62 [1.87, 3.37] <0.01 R2 = 0.66
fully adjusted + risk profile Exam 3 Ref. 0.17 [−0.30, 0.65] n.s 0.90 [0.26, 1.53] <0.01 R2 = 0.76
SAT fully adjusted Ref. 2.60 [1.87, 3.34] <0.01 6.16 [5.23, 7.10] <0.01 R2 = 0.48
fully adjusted + risk profile Exam 1 Ref. 2.05 [1.30, 2.81] <0.01 3.63 [2.56, 4.70] <0.01 R2 = 0.63
fully adjusted + risk profile Exam 3 Ref. 0.22 [−0.26, 0.70] n.s 0.06 [−0.59, 0.71] n.s R2 = 0.87
RSFF fully adjusted Ref. 3.39 [1.02, 5.76] <0.01 3.20 [0.18, 6.22] 0.04 R2 = 0.29
fully adjusted + risk profile Exam 1 Ref. 2.47 [−0.41, 5.35] n.s 0.31 [−3.77, 4.39] n.s R2 = 0.29
fully adjusted + risk profile Exam 3 Ref. 0.19 [−2.80, 3.18] n.s −2.28 [−6.28, 1.72] n.s R2 = 0.33
HFF fully adjusted Ref. 1.54 [1.27, 1.86] <0.01 2.48 [1.93, 3.16] <0.01 R2 = 0.43
fully adjusted + risk profile Exam 1 Ref. 1.51 [1.21, 1.90] <0.01 2.23 [1.62, 3.03] <0.01 R2 = 0.47
fully adjusted + risk profile Exam 3 Ref. 1.14 [0.91, 1.43] n.s 1.32 [0.98, 1.79] n.s R2 = 0.53
PFF fully adjusted Ref. 1.52 [1.26, 1.84] <0.01 2.20 [1.73, 2.80] <0.01 R2 = 0.24
fully adjusted + risk profile Exam 1 Ref. 1.52 [1.21, 1.92] <0.01 1.82 [1.32, 2.51] <0.01 R2 = 0.26
fully adjusted + risk profile Exam 3 Ref. 1.40 [1.10, 1.79] <0.01 1.58 [1.15, 2.20] <0.01 R2 = 0.27

Estimates are derived from a linear regression model. Cluster I served as reference category, i.e. estimates describe the change in adipose tissue outcome that is associated with membership in Cluster II (or Cluster III) as compared to membership in Cluster I. Estimates for TAT, VAT, SAT and RSFF are given as β-coefficients. Estimates for HFF and PFF are back-transformed from log-transformation and are therefore given as %change of the geometric mean. The fully adjusted model is adjusted for age, sex, antihypertensive medication, lipid-lowering medication, smoking status, and validated diabetes.