OBJECTIVE Cross-sectional studies suggest that lipopolysaccharide-binding protein (LBP) may be associated with obesity and metabolic disorders. However, prospective studies examining LBP are lacking. This prospective study investigated the association between LBP and metabolic abnormalities in 580 African ancestry men (mean age, 59.1 ± 10.5 years). RESEARCH DESIGN AND METHODS We measured fasting serum LBP at baseline. Changes in adiposity and glucose homeostasis as well as case subjects with new type 2 diabetes and impaired fasting glucose (IFG) were assessed at a follow-up visit ˜6 years later. Baseline LBP values were tested across quartiles for linear trend with metabolic measures. Multivariable logistic regression was used to determine the odds of new cases of IFG or diabetes per 1-SD greater baseline LBP. RESULTS LBP was significantly associated with baseline BMI, waist circumference, whole-body and trunk fat, skeletal muscle density, fasting serum insulin, and HOMA-insulin resistance (IR) (all P < 0.01). Greater baseline LBP was significantly associated with longitudinal increases in the percentage of trunk fat (P = 0.025) and HOMA-IR (P = 0.034), but only borderline so with a decrease in skeletal muscle density (P = 0.057). In men with normal glucose, baseline LBP was associated with increased odds of having IFG at follow-up after adjustment for age, baseline trunk fat, and lifestyle factors (odds ratio per 1-SD LBP: 1.51; 95% CI 1.02–2.21). This association was attenuated after additional adjustment for change in trunk fat (P = 0.067). CONCLUSIONS LBP may be a marker of prediabetes. Some of this association appears to be mediated through increased central and ectopic skeletal muscle adiposity.