Obesity is a chronic inflammatory disorder that increases the risk of cardiovascular diseases (CVDs). Given the high CVD mortality rate among individuals with obesity, early screening should be considered. Plasminogen activator inhibitor (PAI-1), a cytokine that links obesity and CVDs, represents a promising biomarker. However, PAI-1 is not part of the clinical routine due to its high cost. Therefore, it is necessary to find good predictors that would allow an indirect assessment of PAI-1. This study enrolled 47 women with severe obesity (SO). The obtained anthropometric measurements included weight, height, neck (NC), waist (WC), and hip circumference (HC). Blood samples were collected to analyse glucose and lipid profiles, C-reactive protein, liver markers, adiponectin, and PAI-1 (determined by ELISA immunoassay). Homeostasis model assessment-adiponectin (HOMA-AD), homeostasis model assessment of insulin resistance (HOMA-IR), quantitative insulin sensitivity check index (QUICKI), triglyceride-glucose index (TyG), and atherogenic index of plasma (AIP) were calculated. The women were grouped according to PAI-1 levels. The data were analysed using IBM SPSS Statistics, version 21. The significance level for the analysis was set at 5%. Women with SO who have higher levels of PAI-1 have lower values of high-density lipoprotein cholesterol (HDL) (p = 0.037) and QUICKI (0.020) and higher values of HOMA-AD (0.046) and HOMA-IR (0.037). HOMA-IR was demonstrated to be a good predictor of PAI-1 in this sample (B = 0.2791; p = 0.017). HOMA-IR could be used as a predictor of PAI-1 levels, pointing out the relevance of assessing glycaemic parameters for the prevention of CVDs in women with SO.