Abstract Background Prolactin (PRL) exists in different forms in human serum. The predominant form is monomeric PRL, the other forms include dimeric PRL and polymeric PRL, also known as macroprolactin, a biologically inactive form. Macroprolactin can be measured by almost all immunoassays for PRL. This may lead to misdiagnosis and mismanagement of patients. Polyethylene glycol (PEG) precipitation is the most widely used method for detecting macroprolactin, generally employing a cutoff of 40% to distinguish macroprolactinemia from true hyperprolactinemia. This approach lacks specificity, so the reported result may be misinterpreted in cases in which excess macroprolactin occurs simultaneously with supraphysiological concentrations of monomeric prolactin. Nowadays, it is recommended to report results as post-PEG PRL using method-specific post-PEG reference intervals, classifying as macroprolactinemia (false or pseudohyperprolactinemia) those individuals with post- PEG PRL within post-PEG reference intervals and true hyperprolactinemia above the upper limit of the post-PEG reference interval. The objective of this study was to evaluate the prevalence of macroprolactinemia in patients with hyperprolactinemia using two different criteria, a cutoff < 40% in PEG precipitation and a post- PEG PRL within post-PEG assay specific reference range. Methods We analyzed blood samples requesting serum macroprolactin from patients admitted to a private reference clinical laboratory in Rio de Janeiro, Brazil, from 01/01/2019 to 12/31/2023. Anonymized data on laboratory tests was available from a database of the local Laboratory Information System. All patients included had PRL levels above the reference range, males >15.2 ng/mL, females >23.3 ng/mL (ECLIA, Modular, Roche). Results 9423 patients, 80.1% women (W) were evaluated. Men (M) were statistically significantly older than women (p<0.001) (45.6; SD=15.0 vs. 38.9; SD=11.5). Prolactin median levels were significantly higher in women (p<0.001) than men (37.4ng/mL, IQR=26 vs. 23.8ng/mL, IQR=17.1). Macroprolactinemia detection was greater (p<0.001) by post-peg PRL concentration criteria (9.7% - 10.1%W vs. 8.4%M), than by < 40% PRL recovery criteria (5.5% - 5.7%W vs. 4.8%M). Dividing patients by prolactin levels (<50.0; 50.0 to <100.0; 100.0 to <150.0, >150.0 ng/mL, we found macroprolactinemia in 5.2% (5.2%W vs. 5.2%M), 7.7% (8.2%W vs. 2.4%M), 2.8% (2.8%W vs. 2.5%M) and 1.6% (1.3%W vs. 2.6%M) - <40% PRL recovery criteria; and 12.1% (12.8%W vs. 9.8%M), 4.4% (4.8%W vs. 0.6%M), 0.8% (0.8%W vs. 0%M), 0.0% (0%W vs. 0%M) - post-PEG PRL concentration criteria. By the last criteria, no patient with total PRL more than 150 ng/mL had macroprolactinemia. Conclusions The criterion for defining macroprolactinemia has a big impact on the study results and needs to be considered in data interpretation and comparison with other studies. PEG precipitation is an easy and fast screening method for macroprolactin. Its main purpose should be determining whether the bioactive monomeric prolactin concentration is increased which is why the post-PEG PRL with corresponding reference interval is the most suitable way of reporting results.