Mammary Adipocyte Size, Obesity-Related Breast Adipose Tissue Dysfunction, and Breast Cancer: A Systematic Review and Meta-Analysis

Figures

• 

  Figure 1. Flow diagram of study selection according to PRISMA guidelines. The selection process includes four main stages: identification of studies in databases, initial screening, eligibility assessment, and final inclusion in the systematic review. Numbers in parentheses indicate the number of studies at each stage. PRISMA = Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

• 

  Figure 2. Risk of bias assessment for included studies using ROBINS-E tool [14,19,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46]. (A) Individual study assessment across seven domains (D1–D7). (B) Summary distribution across all domains. Green indicates low risk, yellow indicates some concerns, and red indicates high risk of bias. ROBINS-E = Risk Of Bias In Non-randomized Studies of Exposure.

• 

  Figure 3. Forest plot of the correlation between mammary adipocyte diameter and BMI. Individual study Spearman correlation coefficients with 95% confidence intervals and pooled estimate using a random-effects model. The diamond represents the overall effect size [14,19,33,34,35,36,37,38]. Substantial heterogeneity was detected (I2 = 72.8%). BMI = body mass index.

• 

  Figure 4. Funnel plot assessing publication bias in the correlation between mammary adipocyte diameter and BMI [14,19,33,34,35,36,37,38]. Fisher’s Z-transformed correlation coefficients (with Bonett & Wright correction) plotted against standard error [25]. Each point represents one study. The vertical dashed line indicates the pooled Fisher’s Z-transformed correlation coefficient. The distribution appears broadly symmetrical, and Egger’s test showed no significant evidence of publication bias. BMI = body mass index.

• 

  Figure 5. Forest plot of the correlation between mammary adipocyte diameter and crown-like structure density [14,19,34,37,39]. Individual study Spearman correlation coefficients with 95% confidence intervals and pooled estimate using a random-effects model. The diamond represents the overall effect size. No heterogeneity detected (I2 = 0%).

• 

  Figure 6. Funnel plot assessing publication bias in the correlation between mammary adipocyte diameter and crown-like structure density [14,19,34,37,39]. Fisher’s Z-transformed correlation coefficients (with Bonett & Wright correction) plotted against standard error [25]. Each point represents one study. The vertical dashed line indicates the pooled Fisher’s Z-transformed correlation coefficient. The symmetrical distribution suggests no evidence of publication bias.

• 

  Figure 7. Forest plot of mean difference in mammary adipocyte diameter between CLS-positive and CLS-negative mammary tissue [37,38,39]. Individual study mean differences with 95% confidence intervals and pooled estimate using random-effects model. Diamond represents overall effect size. No heterogeneity detected (I2 = 0%). CLS = crown-like structures; MD = mean difference.

• 

  Figure 8. Funnel plot assessing publication bias in studies comparing mammary adipocyte diameter between CLS-positive and CLS-negative mammary tissue [37,38,39]. Each point represents individual study plotted by mean difference against standard error. The vertical dashed line indicates the pooled mean difference. Symmetrical distribution suggests low risk of publication bias. CLS = crown-like structures.

• 

  Figure 9. Forest plot of correlation between mammary adipocyte diameter and aromatase expression [39,41,43]. Individual study Spearman correlation coefficients with 95% confidence intervals and pooled estimate using random-effects model. Diamond represents overall effect size. Considerable heterogeneity was observed (I2 = 82.9%). Pre = premenopausal; Post = postmenopausal.

• 

  Figure 10. Funnel plot assessing publication bias in the correlation between mammary adipocyte diameter and aromatase expression [39,41,43]. Fisher’s Z-transformed correlation coefficients (with Bonett & Wright correction) plotted against standard error [25]. Each point represents one study. The vertical dashed line indicates the pooled Fisher’s Z-transformed correlation coefficient. The symmetrical distribution suggests no evidence of publication bias. Pre = premenopausal; Post = postmenopausal.