Table 1.
Inclusion and exclusion criteria.
Fig 1.
Flow diagram of the selection process.
*Full-text articles only available in languages other than English, Danish, Swedish, Norwegian, Turkish, and Arabic.
Table 2.
Study characteristics.
Table 3.
SMD and associated heterogeneity in a sensitivity analysis of 24 studies comparing adiponectin levels in LBW individuals with NBW controls (including outliers1).
Fig 2.
The association between circulating adiponectin levels and low birth weight as compared to normal birth weight controls after the exclusion of outlier studies.
Forest plot comparing mean circulating adiponectin levels in LBW subjects and NBW controls after outlier removal. The overall effect size is SMD = −0.46 μg/ml [95% CI: −0.57; −0.35], P < 0.0001. The blue diamond represents the overall effect size and confidence interval. Heterogeneity is estimated at I2 = 67% (P < 0.01), assessed using Cochran’s Q and I2 statistics.
LBW, low birth weight; NBW, normal birth weight; SMD, standardized mean difference; SD, standard deviation; CI, confidence interval; I2, heterogeneity.
Seventeen data points are presented for 16 studies, as Ibäñez et al. [38] included two LBW groups compared to one NBW group.
Fig 3.
The association between circulating adiponectin levels and catch-up growth vs. non-catch-up growth.
Forest plot comparing mean circulating adiponectin levels in CU-growth versus non-CU-growth groups across four studies. The overall effect size is SMD = −0.73 [95% CI: −1.41; −0.05], P = 0.04. The blue diamond represents the overall effect size and confidence interval. Heterogeneity is estimated at I2 = 41.3% (P = 0.16) assessed using Cochran’s Q and I2 statistics.
SMD, standardized mean difference; SD, standard deviation; CI, confidence interval; I2, heterogeneity; CU, catch-up growth.
Catch-up growth is defined as either corrected height SDS ≥ 0 [36,37,44] or matched height, weight, and BMI to NBW controls [38].
Table 4.
SMD and associated heterogeneity in a sensitivity analysis of 16 studies comparing adiponectin levels in LBW individuals with NBW controls (excluding outliers1).
Fig 4.
Funnel plot with Egger’s test.
Funnel plot with Egger’s Test was performed to evaluate the publication bias.
Fig 5.
Analysis of pooled Pearson correlations from 19 studies examining the relationship between adiponectin levels and birth weight.
Forest plot showing the correlation between circulating adiponectin levels and birth weight across 19 studies. The pooled estimate is r = 0.31. Heterogeneity is estimated at I2 = 89% (P < 0.0001), assessed using Cochran’s Q and I2 statistics. Pearson correlations (r) were extracted from eligible studies, and 95% CIs were calculated using the random effects model.
SMD, standardized mean difference; CI, confidence interval; I2, heterogeneity.
Twenty data points are presented for 19 studies, as Meral et al. [38] provide a correlation coefficient for both small and large for gestational age.
Fig 6.
The association between circulating adiponectin levels and high birth weight as compared to normal birth weight controls.
Forest plot comparing mean circulating adiponectin levels in HBW versus NBW groups. The overall effect size is SMD = −0.36 μg/ml [95% CI: −1.06; 0.33], P = 0.27. The blue diamond represents the overall effect size and confidence interval. Heterogeneity is estimated at I2 = 95% (P < 0.01), assessed by Cochran’s Q and I2 statistics.
NBW, normal birth weight; HBW, high birth weight; SMD, standardized mean difference; SD, standard deviation; CI, confidence interval; I2, heterogeneity.