Concordance of Capillary Electrophoresis and Conventional Gel Electrophoresis in Two Different Groups of Patients with.

BACKGROUND: Serum and urinary protein electrophoresis play an important role in the identification of monoclonal gammopathy. Recently, capillary electrophoresis (CE) has been adapted in many clinical laboratories because of several advantages such as short turnaround time, automation, and high reproducibility. However, there have been unsolved concerns for the concordance between conventional gel and automated capillary electrophoresis methods for protein separation in clinical specimens. In this study, we investigated the diagnostic performance of both methods for detecting monoclonal (M) protein.

METHODS: From February 2012 to August 2015, a total of 3,013 CE tests were performed in our hospital. Among these cases, we reconfirmed results of CE (Capillary 2, Sebia, Lysse, France) with those of conventional agarose gel electrophoresis (GE) (Hydragel 4IF, Sebia, Lisses, France) in 28 specimens from 24 patients with newly diagnosed monoclonal gammopathy (group 1). In addition, 22 cases from 15 patients with previously diagnosed monoclonal gammopathy presenting indeterminate or suspicious results on CE (group 2) were also reconfirmed with GE.

RESULTS: We compared the results between the two electrophoresis methods in two different groups of patients with newly diagnosed discrete monoclonal peaks vs. pre-existing monoclonal gammopathy with obscure results in follow-up courses. In group 1, agreement rate was 100% (28/28) and there was no discrepant result between these two electrophoresis methods. In contrast, group 2 showed 86.4% (19/22) agreement rate and 0.67 Cohen's kappa value (95% confidence interval, 0.51 - 1.02).

CONCLUSIONS: According to our results, both electrophoresis methods can be used with the same level of assurance at the time of initial diagnosis for monoclonal gammopathy. However, in patients with previously diagnosed monoclonal gammopathy in follow-up course after appropriate treatments, discordant results can be observed due to the reduced amount of M proteins. Therefore, we suggest that some ambiguous cases with very small amounts of M components require a combination of both CE and GE methods for accurate interpretation to confirm the presence of M proteins.