Abstract Background Serum creatinine test is used to diagnose and monitor the progression of renal diseases. Accuracy and precision are very important for this test. Enzymatic creatinine is often considered as the preferred method due to its accuracy and less interference from bilirubin. However, inaccurate results with enzymatic assay could still occur in some cases due to the interference by the M-protein. High M-protein in blood often occur in patients with multiple myeloma or rheumatoid diseases. This interference is often obscured in testing because the appearances of these samples are not significantly different from normal samples. There are serveral IVD companies who provide enzymatic creatinine reagents. Not all of them resolved M-protein interference. This study systematically investigated the causes and patterns of interferences by M-protein to our current creatinine reagent. Based on the findings, we improved reagent formulation to prevent such interference. Methods This assay is based on the sarcosine oxidase creatinine method with dual-reagents on BS-2800M. Possible M-protein interferences to the current assay were explored with 1588 serum samples collected from local hospitals. These samples either contain high concentrations of globulins or from myeloma or rheumatoid diseases patients. These samples were tested for creatinine using reagents from several manufacturers. Discrepancies in creatinine results were found in 386 samples among the assays with these reagents. The reaction kinetics of our creatinine assay from these samples were evaluated in order to identify the samples with M-protein interference and the characteristics of this interference. Thirty-eight samples were identified to have M-protein interference. Based on the findings, new reagent formulation was developed. To confirm the effectiveness of the anti M-protein interference, we compared the results of the most sever M-protein interference samples with improved reagent and an in-house ID-LC-MS/MS creatinine reference method. The reference method was calibrated with SRM 914a and validated by SRM 967a. Results With M-protein interference samples, there were three patterns found in reaction kinetics, rises, drops or unstable in certain periods vs those with similar creatinine concentrations from normal samples. All these abnormalities in reaction kinetics resulted in inaccurate results. We further investigated these samples by measuring albumin and globulins as well as precipitating proteins and measuring supernatant. We found that samples with more severe interference also had higher IgM concentrations. In the supernatant, the consistence of the creatinine concentrations was obtained between corresponding creatinine tests and the LC-MS/MS measurements. Within 38 M-protein interference samples identified, 13 samples shown the most significant interference. By using the improved creatinine reagent, these samples shown consistent results vs. the results from the in-house LC-MS/MS reference method. Conclusion We studied the causes of M-protein interference to creatinine assay and analyzed the characteristics of such interference. Based on the findings, we improved our reagent formulation with enhanced anti M-protein interference capability. The comparison with LC-MS/MS indicated that Mindray new reagent has high consistency with the results for the samples identified with M-protein interference vs. the reference method. This new creatinine reagent can be used in clinical laboratories to avoid interference from M-proteins.
Wang et al. (Wed,) studied this question.