Time-varying and tissue-dependent effects of adiposity on leptin levels: A Mendelian randomization study
Figures

Forest plots and schematic diagrams for Lifecourse Mendelian randomization analysis.
(A) A forest plot illustrating the direct effect of childhood body size on circulating leptin levels measured during childhood (mean age: 9.9 y) using individual-level data from the Avon Longitudinal Study of Parents and Children (ALSPAC). (B) provides the corresponding schematic diagram for ALSPAC results. (C) A forest plot depicting the indirect effect of childhood body size on adulthood measured leptin levels using data from the deCODE Health study (mean age: 55 y) as portrayed in the schematic diagram presented in panel (D). ‘Genetic variants’ refers to instruments for both exposures in the model. The red arrow indicates the causal pathway being assessed. MR, Mendelian randomization. Data underlying this figure can be found in Supplementary file 1b and d.

Forest plots and schematic diagrams for Tissue-partitioned Mendelian randomization analysis.
(A) A forest plot illustrating the attenuation of the adipose tissue-instrumented body mass index (BMI) effect on childhood leptin levels, whereas estimates for brain tissue-instrumented BMI remained robust. (B) The corresponding schematic diagram for this finding. (C) A forest plot displaying similar conclusions to the estimates found in childhood, but derived using adulthood measured leptin levels. (D) provides the corresponding schematic diagram for estimates portrayed in panel (C). ‘Genetic variants’ refers to instruments for both exposures in the model. The red arrow indicates the causal pathway being assessed. MR, Mendelian randomization. Data underlying this figure can be found in Supplementary file 1h and j.

Forest plots depicting the univariable and multivariable Mendelian randomization (MR) effect estimates of adipose tissue-instrumented body mass index (BMI) (yellow) and brain tissue-instrumented BMI (red) on adulthood measures of (A) abdominal subcutaneous fat volume, (B) visceral adipose fat volume, and (C) gluteofemoral fat volume.
Data underlying this figure can be found in Supplementary file 1m.
Tables
Study | Exposure | Outcome | Β | SE | P |
---|---|---|---|---|---|
deCODE | Circulating leptin | BMI | 0.024 | 0.039 | 0.541 |
Additional files
-
Supplementary file 1
Supplementary tables.
(a) Exposures and outcomes in this study. (b) Multivariable Mendelian randomization using lifecourse instruments on childhood leptin. (c) Univariable Mendelian randomization using lifecourse instruments. (d) Multivariable Mendelian randomization using lifecourse instruments on adult leptin. (e) Univariable and multivariable lifecourse Mendelian randomization analyses using adulthood leptin data derived from deCODE. (f) Multivariable Mendelian randomization analysis of DXA-derived measures in ALSPAC. (g) Multivariable Mendelian randomization analysis of skinfold measures in ALSPAC. (h) Multivariable Mendelian randomization using tissue-partitioned instruments on childhood leptin. (i) Univariable Mendelian randomization using tissue-partitioned instruments. (j) Multivariable Mendelian randomization using tissue-partitioned instruments on adult leptin. (k) Univariable and multivariable tissue-partitioned Mendelian randomization analyses using adulthood leptin data derived from deCODE. (l) Cochran’s Q-statistics and MR-Egger intercept terms. (m) Multivariable Mendelian randomization using tissue-partitioned instruments on adult measures of fat distribution.
- https://cdn.elifesciences.org/articles/84646/elife-84646-supp1-v1.xlsx
-
MDAR checklist
- https://cdn.elifesciences.org/articles/84646/elife-84646-mdarchecklist1-v1.docx