Table 2 shows the adipokine concentrations and adipokine-derived ratios (leptin, total adiponectin, HMWA, LAR, and leptin/HMWA ratio).
One possible way to evaluate the performance of plasma leptin, total adiponectin, HMWA, LAR, and leptin/HMWA ratio with insulin sensitivity indexes is by comparing their slopes, where a greater slope means a stronger magnitude of association.
We additionally evaluated such associations using residual analysis, finding that associations between studentized residuals of leptin, total adiponectin, HMWA, LAR, and leptin/HMWA ratio (all log transformed) with respect to BMI showed significant associations with indexes of insulin sensitivity.
Likewise, no significant differences were found in ROC curve areas for plasma leptin, total adiponectin, and HMWA in relation to insulin sensitivity indexes (Table S2).
We have confirmed that plasma leptin levels were positively associated with BMI and negatively with insulin sensitivity indexes, whereas the opposite pattern of associations was observed for total adiponectin and HMWA. Such well-known associations emphasize the importance of these two adipokines connecting adipose tissue biology, obesity, and insulin sensitivity.
We speculate that the predominant effect of leptin/adiponectin ratios (either LAR or leptin/HMWA ratio) on insulin sensitivity is apparently driven by leptin over adiponectin (or HMWA), given that the associations of leptin/adiponectin with HOMA-S, Matsuda-ISICOMP, or CSi display a negative slope (the same as what occurs with plasma leptin).
We have confirmed that both adiponectin and HMWA are positively associated with insulin sensitivity indexes and negatively with BMI, even in our group of normoglycemic, mainly normal-weight women with no family history of diabetes.
We have not found publications evaluating the specific influence of including HMWA instead of total adiponectin on the performance of leptin/adiponectin ratios as biomarkers of systemic (whole body) insulin sensitivity.
We have confirmed in our group of normoglycemic women with no family history of diabetes that plasma leptin levels were positively associated with BMI and negatively with insulin sensitivity indexes, whereas the opposite was found both for adiponectin and HMWA. LAR showed consistent negative associations with different systemic insulin sensitivity indexes derived from basal samples, as well as, from oral or intravenous glucose loads.
In previous studies, the extensive amino acid diversity and the lack of disease-specific hmwA sequence clusters underscore the challenges of targeting HMW adhesins as vaccine components 2 but in this study, specific sequences of hmwA in the specific sites were observed.
Prevalence, distribution, and sequence diversity of hmwA among commensal and otitis media non-typeable Haemophilus influenzae.
Identification of new hmwA alleles from nontypeable Haemophilus influenzae.