Comment on Hedderson et al. Prepregnancy SHBG Concentrations and Risk for Subsequently Developing Gestational Diabetes Mellitus. Diabetes Care 2014;37:1296–1303

Hedderson et al. (1) recently found that lower sex hormone–binding globulin (SHBG) concentrations measured, on average, 6 years before pregnancy were associated with increased odds of the development of gestational diabetes mellitus (GDM). Because this relationship was independent of known risk factors of GDM, they concluded that circulating SHBG may represent a potentially useful new biomarker for identifying who is at risk for GDM beyond the currently established clinical and demographic risk factors, in particular insulin resistance.

While the study appears to be well performed and the data are very interesting, the conclusion with respect to insulin resistance may be overstated. In their study, Hedderson et al. could estimate insulin resistance only from fasting glucose and insulin levels by calculating the HOMA-IR. A dynamic test, e.g., a frequently sampled 75-g oral glucose tolerance test, allows more precise determination of insulin resistance. We have previously reported on relationships of circulating SHBG with insulin resistance and precisely measured body fat composition and liver fat content (2,3). In women (N = 135, age 45 ± 1 years [means ± SE]), we observe stronger correlations of circulating SHBG with insulin sensitivity estimated from the index proposed by Matsuda and DeFronzo [10,000/√(Ins_mean ⋅ Gluc_mean ⋅ Ins₀ ⋅ Gluc₀)] (r = 0.23, P = 0.007) than with insulin resistance estimated by the HOMA-IR (r = −0.18, P = 0.03). Thus, low circulating SHBG may reflect higher insulin resistance, which may strongly predict GDM also in the study by Hedderson et al. (1). Nevertheless, if replicated in other studies, circulating SHBG may represent an interesting novel biomarker in fasting blood samples for incidence of GDM.

In addition, low circulating SHBG may reflect increased amount of liver fat content. In our studies, circulating SHBG correlated stronger with liver fat content, both in cross-sectional and longitudinal analyses, than with total body and visceral fat mass or whole-body insulin sensitivity (2,3). Interestingly in their most recent article, Hedderson and colleagues (4) found in the same study that increased γ-glutamyl transferase levels were associated with increased odds of the development of GDM. In that article, they concluded that this may represent higher liver fat content and hepatic insulin resistance. It would have been interesting to know whether γ-glutamyl transferase levels have impact on the previously observed relationship of circulating SHBG with the development of GDM (1).

Hedderson et al. (1) also addressed lifestyle factors that may affect circulating SHBG in women. In this respect, they highlighted that loss of body weight induced by changes in diet and physical activity may result in an increase of circulating SHBG. In this respect, we showed that loss of liver fat content most strongly correlated with increase in circulating SHBG during a lifestyle intervention (2,3). This is in agreement with published information indicating that reduction in hepatic inflammatory signaling and hepatic lipogenesis mediates this increase (5).

In summary, the data by Hedderson et al. (1) add interesting novel information about the role of the liver-derived protein SHBG in the development of GDM. In addition, their data support the novel concept that hepatokines are very important for the regulation of metabolism (5).