Obesity (Silver Spring) 2017; 25:2092-2099. doi:10.1002/oby.22025 The authors of this article recently discovered that the C-reactive protein values used in the analysis were approximately threefold higher than the true concentration. The manufacturers of the ELISA kit recently informed the authors that the kit instruction manual gave the wrong volume of buffer needed to reconstitute the standard. This laboratory error affected the data presented in Tables and and Figure of the manuscript. After rerunning all of the statistical analyses with the corrected C-reactive protein values, there are no changes in either the statistical significance of the associations or the conclusions and interpretation of the findings. The authors apologize for this error. Summary statistics for breast milk inflammatory markers at 1 and 3 months post partum (Table presented.) Data presented as median (interquartile range) or mean ± SD. aP value testing for differences between 1- and 3-month summary statistics using Wilcoxon signed rank sum test for nontransformed data and paired t tests for log-transformed data. Bolded values are statistically significant (P<0.05). bThe value of 1 was added to the raw interleukin-6 values before log transformations were performed to avoid negative values on the log scale. Associations of maternal anthropometrics with log-transformed breast milk C-reactive protein at 1 and 3 months post partuma (Table presented.) aAll continuous independent variables (pre-pregnancy BMI, gestational weight gain, and postpartum weight loss) were standardized to a mean of 0 and a standard deviation of 1. Bolded values are statistically significant (P<0.05). bModel 1 is the crude model. cModel 2 adjusts for maternal education and gestational age at delivery. dModel 3 adjusts for covariates in Model 2, as well as pre-pregnancy BMI. eCalculated using the 2009 Institute of Medicine gestational weight gain guidelines, based on pre-pregnancy BMI. fPostpartum weight loss was calculated as maternal weight at delivery minus maternal weight at 1 and 3 months post partum and was entered into the mixed effects models as a time-varying independent variable. (Figure presented.) Adjusted means of back-transformed C-reactive protein (CRP) at 1 and 3 months post partum by pre-pregnancy body mass index (BMI) and gestational weight gain (GWG) category. Differences between the mean values for each pre-pregnancy BMI and GWG category are denoted by different superscript letters (a, b, c, or d). In the top panel, at 1 month post partum, women with normal weight who gained weight below/within the guidelines (a) had significantly lower breast milk CRP than all other groups (b), except overweight women who gained weight below/within the guidelines. Women with obesity who gained below/within GWG guidelines (c) had significantly higher CRP than women with overweight, regardless of GWG category, and women with normal weight who gained below/within GWG guidelines (d). In the bottom panel, at 3 months post partum, women with normal weight who gained weight below/within guidelines (a) had significantly lower CRP than women with normal weight who exceeded GWG guidelines and women with obesity, regardless of GWG category (b). Women with obesity who experienced excessive weight gain (c) had significantly higher CRP than all normal weight and overweight groups (d; P<0.05 for all).
Bibliographical notePublisher Copyright:
© 2018 The Obesity Society