Evaluation of two fully automated novel enzyme-linked immunosorbent assays for the determination of human adiponectin in serum

Risch, Lorenz; Hoefle, Guenter; Saely, Christoph; Berchthold, Sabine; Weber, Martina; Gouya, Ghazaleh; Rein, Philipp; Langer, Peter; Marte, Thomas; Aczel, Stefan; Drexel, Heinz (2006). Evaluation of two fully automated novel enzyme-linked immunosorbent assays for the determination of human adiponectin in serum. Clinica chimica acta, 373(1-2), pp. 121-6. Amsterdam: Elsevier 10.1016/j.cca.2006.05.017

Full text not available from this repository. (Request a copy)

BACKGROUND: In contrast to RIA, recently available ELISAs provide the potential for fully automated analysis of adiponectin. To date, studies reporting on the diagnostic characteristics of ELISAs and investigating on the relationship between ELISA- and RIA-based methods are rare. METHODS: Thus, we established and evaluated a fully automated platform (BEP 2000; Dade-Behring, Switzerland) for determination of adiponectin levels in serum by two different ELISA methods (competitive human adiponectin ELISA; high sensitivity human adiponectin sandwich ELISA; both Biovendor, Czech Republic). Further, as a reference method, we also employed a human adiponectin RIA (Linco Research, USA). Samples from 150 patients routinely presenting to our cardiology unit were tested. RESULTS: ELISA measurements could be accomplished in less than 3 h, measurement of RIA had a duration of 24 h. The ELISAs were evaluated for precision, analytical sensitivity and specificity, linearity on dilution and spiking recovery. In the investigated patients, type 2 diabetes, higher age and male gender were significantly associated with lower serum adiponectin concentrations. Correlations between the ELISA methods and the RIA were strong (competitive ELISA, r=0.82; sandwich ELISA, r=0.92; both p<0.001). However, Deming regression and Bland-Altman analysis indicated lack of agreement of the 3 methods preventing direct comparison of results. The equations of the regression lines are: Competitive ELISA=1.48 x RIA-0.88; High sensitivity sandwich ELISA=0.77 x RIA+1.01. CONCLUSIONS: Fully automated measurement of adiponectin by ELISA is feasible and substantially more rapid than RIA. The investigated ELISA test systems seem to exhibit analytical characteristics allowing for clinical application. In addition, there is a strong correlation between the ELISA methods and RIA. These findings might promote a more widespread use of adiponectin measurements in clinical research.

Item Type:

Journal Article (Original Article)


04 Faculty of Medicine > Department of Dermatology, Urology, Rheumatology, Nephrology, Osteoporosis (DURN) > Clinic of Rheumatology, Clinical Immunology and Allergology

UniBE Contributor:

Höfle, Günter Adolf










Factscience Import

Date Deposited:

04 Oct 2013 14:52

Last Modified:

05 Dec 2022 14:16

Publisher DOI:


PubMed ID:


Web of Science ID:



https://boris.unibe.ch/id/eprint/21986 (FactScience: 21342)

Actions (login required)

Edit item Edit item
Provide Feedback