Table 2.
Quartiles of ceramide species concentration * | P trend | HR per 1 SD increment † | P value | ||||
---|---|---|---|---|---|---|---|
| |||||||
Q1 | Q2 | Q3 | Q4 | ||||
Ceramide (16:0) | |||||||
Cases | 38 | 57 | 57 | 78 | |||
MV1 ‡ | Ref. | 1.60 (1.00, 2.54) | 1.67 (1.04, 2.67) | 2.20 (1.40, 3.46) | <.001 | 1.43 (1.20, 1.70) | <.001 |
MV2 § | Ref. | 1.72 (1.05, 2.81) | 1.87 (1.14, 3.07) | 2.39 (1.49, 3.83) | <.001 | 1.42 (1.19, 1.69) | <.001 |
Ceramide (22:0) | |||||||
Cases | 53 | 43 | 62 | 72 | |||
MV1 | Ref. | 0.95 (0.60, 1.50) | 1.29 (0.83, 1.99) | 1.89 (1.22, 2.93) | 0.002 | 1.33 (1.12, 1.58) | 0.001 |
MV2 | Ref. | 0.88 (0.54, 1.43) | 1.28 (0.81, 2.02) | 1.91 (1.21, 3.01) | 0.003 | 1.32 (1.10, 1.57) | 0.002 |
Ceramide (24:0) | |||||||
Cases | 48 | 56 | 59 | 67 | |||
MV1 | Ref. | 1.26 (0.80, 1.97) | 1.40 (0.89, 2.18) | 1.88 (1.20, 2.95) | 0.006 | 1.29 (1.09, 1.53) | 0.003 |
MV2 | Ref. | 1.20 (0.75, 1.94) | 1.51 (0.94, 2.42) | 1.97 (1.21, 3.20) | 0.004 | 1.32 (1.10, 1.57) | 0.002 |
Ceramide (24:1) | |||||||
Cases | 44 | 50 | 59 | 77 | |||
MV1 | Ref. | 1.07 (0.67, 1.70) | 1.31 (0.82, 2.08) | 1.53 (0.98, 2.37) | 0.037 | 1.22 (1.04, 1.43) | 0.015 |
MV2 | Ref. | 1.16 (0.72, 1.89) | 1.44 (0.88, 2.36) | 1.73 (1.09, 2.74) | 0.011 | 1.27 (1.08, 1.49) | 0.004 |
Ceramide score | |||||||
Cases | 45 | 51 | 59 | 75 | |||
MV1 | Ref. | 1.14 (0.72, 1.81) | 1.53 (0.97, 2.41) | 2.04 (1.30, 3.18) | <.001 | 1.40 (1.17, 1.66) | <.001 |
MV2 | Ref. | 1.25 (0.77, 2.03) | 1.68 (1.05, 2.69) | 2.18 (1.36, 3.49) | <.001 | 1.41 (1.17, 1.68) | <.001 |
Abbreviations: MV, multivariable model
Quartiles were calculated based on the distribution of the ceramide concentrations in the sub-cohort.
A logarithmic transformation was applied to the raw value.
Model 1 stratified on intervention group and simultaneously adjusted for age (continuous) and sex (male, female).
Model 2 additionally adjusted for body mass index (kg/m2, continuous), family history of premature coronary heart disease (yes, no), smoking status (current, never, former), histories of hypertension, dyslipidemia, and diabetes (all yes, no).