The short-term and long-term effects of bariatric/metabolic surgery on subcutaneous adipose tissue inflammation in humans

Both the short- and longer-term metabolic improvements following bariatric/metabolic surgery occur without significant reductions in measures of adipose tissue inflammation, as assessed by measuring the expression of genes encoding key mediators of inflammation and by flow cytometric immunophenotyping and quantification of adipose tissue leukocytes.

To investigate whether a reduction in adipose tissue inflammation plays a role in the metabolic improvements seen after bariatric/metabolic surgery, both in the short-term and longer-term. Design: Fasting blood and subcutaneous abdominal adipose tissue were obtained before (n=14), at one month (n=9), and 6-12months (n=14) after bariatric/metabolic surgery from individuals with obesity who were not on insulin or anti-diabetes medication. Adipose tissue inflammation was assessed by a combination of whole-tissue gene expression and flow cytometry-based quantification of tissue leukocytes.

One month after surgery, body weight was reduced by 13.5±4.4kg (p<0.001), with improvements in glucose tolerance reflected by a decrease in area-under-the-curve (AUC) glucose in 3-h oral glucose tolerance tests (-105±98mmol/L * min; p=0.009) and enhanced pancreatic β-cell function (insulinogenic index: +0.8±0.9pmol/mmol; p=0.032), but no change in estimated insulin sensitivity (Matsuda insulin sensitivity index [ISI]; p=0.720). Furthermore, although biomarkers of systemic inflammation and pro-inflammatory gene expression in adipose tissue remained unchanged, the number of neutrophils increased in adipose tissue 15-20 fold (p<0.001), with less substantial increases in other leukocyte populations. By the 6-12month follow-up visit, body weight was reduced by 34.8±10.8kg (p<0.001) relative to baseline, and glucose tolerance was further improved (AUC glucose -276±229; p<0.001) along with estimated insulin sensitivity (Matsuda ISI: +4.6±3.2; p<0.001). In addition, improvements in systemic inflammation were reflected by reductions in circulating C-reactive protein (CRP; -2.0±5.3mg/dL; p=0.002), and increased serum adiponectin (+1358±1406pg/mL; p=0.003). However, leukocyte infiltration of adipose tissue remained elevated relative to baseline, with pro-inflammatory cytokine mRNA expression unchanged, while adiponectin mRNA expression trended downward (p=0.069).