The aim of this study was to verify if the selenium status of patients residing in locations with selenium-poor soil who receive parenteral nutrition (PN) without selenium supplementation is associated with the inflammatory process. This was a prospective cohort study with hospitalized patients who started PN. The analyzed biochemical tests were plasma selenium, glutathione peroxidase (GPx), C-reactive protein, prealbumin, albumin, creatinine, lymphocytes, total cholesterol, high-density lipoprotein, and triglycerides. Seventy-seven patients with a mean age of 56.2 ± 15.7 y were studied. Most of them used PN as a result of clinical issues (70.1%) such as, gastric, renal, or hematologic neoplasia; gastrointestinal dysfunction; pancreatitis; sepsis; trauma without surgical needs; chylothorax; and fistula not related to surgical procedure. There were low levels of plasma selenium (98.7%) and GPx (60%) and elevated C-reactive protein (98.5%) in most cases. At the beginning of PN there was no correlation between selenium and laboratory tests (P > 0.05). At the second evaluation (seventh day of PN), there was a positive correlation of selenium levels with lymphocyte levels (r = 0.36; P = 0.04). After 2 wk of PN, there was a statistically significant correlation between selenium and GPx (r = 0.70; P = 0.02). Very low values of selenium and GPx from the beginning of PN were identified. The correlation of selenium levels with GPx in only 14 d of PN, regardless of inflammation, may reflect a critical selenium status, mainly because the correlation was verified after the acute phase. Therefore it is important to emphasize that supplementation should be started from the beginning of PN, especially in regions with selenium-deficient soil.