Usefulness of surrogate markers to determine insulin action in fat cells

Abstract

Obesity is a major factor behind insulin resistance. The validity of simple biochemical surrogate measures to estimate insulin resistance at the fat cell level is unclear. To investigate if the surrogate measures HOMA-IR (glucose/insulin product) and Adipo-IR (fatty acids/insulin product) reflect insulin action on glucose/lipid metabolism in fat cells. Insulin-induced lipogenesis and lipolysis inhibition (antilipolysis) in subcutaneous fat cells were investigated for sensitivity (reflecting receptor-near events) and responsiveness (i.e., maximum action reflecting distal post-receptor events) in 363 subjects. Results were compared with log10 transformed values for HOMA-IR and Adipo-IR. Individually, the four measures of in vitro insulin action on fat cells correlated significantly (p < 0.0001) but weakly with each other (adjusted r2 0.05-0.23). HOMA-IR and Adipo-IR correlated strongly with each other (adjusted r2 = 0.81). Using Spearman or simple linear regression all in vitro measures except antilipolytic responsiveness expressed per lipid weight, correlated significantly with Adipo-IR or HOMA-IR (p values <0.0001). Similar relationships remained after combined correction for age, body mass index and sex. Together, the four in vitro measures explained 50% of the variability in HOMA-IR and ADIPO-IR (p < 0.0001). Receiver-operating characteristic analysis showed good sensitivity and specificity for Adipo-IR and HOMA-IR to detect combined insulin resistance of antilipolysis and lipogenesis in fat cells (area under the curve = 0.8). Insulin action at the receptor and post-receptor levels on lipolysis and lipogenesis in fat cells correlates significantly with Adipo-IR and HOMA-IR. Both surrogate measures give similar information about insulin resistance of glucose and lipid metabolism in fat cells.