Systemic Adiponectin Values in Humans Require Standardized Units

To the editor: We took great interest in reading the article by Dal Molin Netto et al. [1], where they observed the increase of antiinflammatory adipokines, like adiponectin after Roux-en-Y gastric bypass. This is an interesting conclusion that was based in part on adiponectin levels and leptin/adiponectin ratio. Unfortunately, this result could mislead the reader since adiponectin levels that are presented in ng/ml in this study are 1000 times below than the expected classical values in such studied subjects. Adiponectin is a highly studied adipokine in many scientific and medical areas and is currently the subject of more than 14,000 Medline citations. Human adiponectinemia usually ranges from 2 to 20 μg/mL or mg/L [2]. However, in the literature, different units are commonly used to express adiponectinemia, such as ng/mL, μg/mL, mg/dL, or mg/L. This lack of consensus on how to express adiponectin circulating levels leads to honest mistakes and potential misinterpretation of the results. Thus, even if the absolute value of adiponectinemia is correct, in some studies the units are not appropriate to this value, which could confuse the reader. This is exactly what it is observed in the study from Dal Molin Netto et al. [1]. To further illustrate this, if we have an adiponectinemia of 5 mg/L, this equals 0.5 mg/dL, 5 μg/mL, or 5,000 ng/mL, but is not the same as 5 mg/dL or 5 ng/mL or 50 ng/mL like it was suggested, for example, by several recently published studies [1, 3–6]. These few examples of Berroneous^ circulating adiponectin values can be misleading. This mistake could result from the fact that, in many adiponectin ELISA or RIA assays, the values range from one to several hundred ng/mL and serum or plasma samples need to be diluted before the assay. Thus, the results must be multiplied by the dilution factor to obtain the right value of adiponectin expressed in μg/mL or mg/L. Sometimes the conversion is made for the value without changing the units, i.e., conversion of ng/mL into μg/mL or mg/L, which could explain the errors in many cases. Two recent studies aimed to established reference values for adiponectin in children and adolescents, which emphasize the major importance to use standardized units to express circulating adiponectin values [7, 8]. In conclusion, in order to avoid any confusion, we suggest the systematic use of μg/mL or mg/L to express circulating adiponectin levels in humans, to obtain * Jean-Philippe Bastard jean-philippe.bastard@aphp.fr