To diagnose invasive fungal infections, the detection of (1 → 3)-β- d -glucan in serum has shown variable specificity, depending on the targeted population. Several circumstances for false-positive results of beta-glucan tests have been identified, among which are severe bacterial infections. In this study, we measured (1 → 3)-β- d -glucan by the Fungitell test in the serum of 62 patients (one serum sample tested per patient) for whom invasive fungal infection was not suspected: 19 control subjects and 43 patients with bacteraemia. The test was interpretable for 58 sera: all 19 control subjects had negative beta-glucan test; among the 39 bacteraemic patients, we report 16 false-positive results. For the 22 patients undergoing bacteraemia due to Gram-negative bacilli, we observed 13 false-positive results (59%). Among the 17 patients with bloodstream infection involving Gram-positive cocci, three false-positive tests were recorded, but none in the eight cases of Streptococcus pneumoniae bacteraemia. Statistical analysis showed that beta-glucan levels were significantly higher in patients with Gram-negative bacilli bloodstream infection in comparison to those with bacteraemia due to Gram-positive cocci. These results were independent from other previously described causes for false-positive beta-glucan tests. These data might help physicians to interpret positive beta-glucan detection when an invasive fungal infection is suspected, especially for patients with bacterial infections.