Reference Interval for Lactate Dehydrogenase Catalytic Activity in Serum Measured According to the New IFCC Recommendations

The present efforts to standardize the measurements of catalytic concentrations of enzymes are directed to the achievement of comparable test results for human serum samples, independently of the measurement test kits and instruments and of the laboratory where the procedure is carried out, and, consequently, to the definition of common reference intervals to avoid confusion among patients and practitioners when results are clinically interpreted (1). In order to achieve this goal, an approach is needed that would provide a reliable transfer of the measurement values of an uppermost recommended reference procedure to methods that are routinely used in the laboratories. Such a structure is presented by the reference system, based on the concepts of metrological traceability and of hierarchy of analytical measurement procedures (2). Since the accuracy of an enzyme reference system resides in the reference measurement procedure, IFCC has recently established new reference procedures for the measurement of creatine kinase, lactate dehydrogenase (LD), alanine aminotransferase, aspartate aminotransferase and γ-glutamyltransferase catalytic activities (3–7). The original IFCC-recommended methods have been modified to optimize them at 37 °C and, at least in some cases, this method optimization has led to a significant increase in the measured enzyme activities (8). The former reference limits can therefore no longer be valid. In this study, we have determined the reference interval for LD activity in serum in accordance with the new IFCC reference measurement procedure (4). LD activities were measured with a Dade Behring Mega® analyzer (Liederbach, Germany), using the IFCCrecommended measurement conditions and homemade reagents prepared as previously described (4). This automatic system was preliminarily validated by direct comparison with the IFCC LD reference measurement procedure performed by the use of a manually operated Lambda 2 spectrophotometer (Perkin-Elmer, Überlingen, Germany) and self-prepared reagent solutions in our IFCC reference laboratory, using the certified reference material (CRM) 453 from the Institute for Reference Materials and Measurements (IRMM) for trueness control. In particular, 50 fresh (leftover) human serum samples were assayed in duplicate with both methods across four different runs. The observed LD activities were distributed across the reportable range for assay (107 to 589 U/l), using protocols from the National Committee for Clinical Laboratory Standards (NCCLS) EP9-A2 document for sample distribution selection and statistical evaluation of data (9). Linear regression analysis gave the following results: y (Mega) = 0.999 (±0.002) × – 0.11 (±0.78), Sy|x = 2.6 U/l, r = 0.9999. Evaluating the intermethod bias, the mean differences (absolute and in percentage) were –0.3 U/l [95% confidence interval (CI): –1.0/+0.4] and –0.10% (95% CI: –0.37/+0.17), respectively, showing that the automatic method was able to provide results traceable to the IFCC reference system for LD. All Mega runs were also validated by the CRM 453 results [mean ± SD: 503.8 U/l ± 9.7, CV = 1.93%; IRMM certified concentration (uncertainty): 502 U/l (7.0 U/l)]. Serum specimens from 157 apparently healthy adult individuals (135 men and 22 women, aged 18– 66 years) were then analyzed. The subjects were drawn from a population of Caucasian blood donors and selected a priori using the IFCC recommendations for the