Drug Information Journal - March 2009 - (Page 149) Local Laboratory Reference Intervals MEDICAL INFORMATION 149 considered here—particularly when compared with the variability seen in the local laboratory reference intervals. These data reinforce the argument that local laboratory reference intervals do not reflect true biological variability between the human populations served by those laboratories. DISCUSSION AND CONCLUSION The analysis clearly demonstrates that local reference intervals are highly variable without a clear relation to differences in measurement bias due to analytical methods. Since we observe these differences after a harmonization step for routine clinical biochemistry analytes, this observation is not related to method differences between laboratories, as is frequently argued in the literature (1,3). Consequently, it is believed that the use of different local reference intervals for cross-study population analyses (eg, to identify toxicity trends) is highly challenging. Data harmonization to allow analysis with a common reference interval is attractive, but method differences in routine clinical biochemistry may create considerable noise in such analysis, especially for analytes with a wide range of currently applied methodology such as alkaline phosphatase and lactate dehydrogenase (13). In consequence, the application of a common reference interval for routine safety analytes is recommended, after method differences in clinical biochemistry are corrected by a harmonization step. Such harmonization is more practical and reliable than the use of external quality assessment or proficiency testing data as proposed by Chuang-Stein (14). The constancy of the method bias of the laboratory can be monitored by regular stability checks throughout the duration of a study. In addition to the use of a common reference interval, results for an individual can also be assessed in relation to their baseline values. Chuang-Stein has proposed several interesting metrics for that purpose, but without a clear guideline as to the identification of an unusual result (14). Fraser et al. have proposed such guidelines, based on analytical imprecision and within-subject bio- logical variation and the calculated reference change value (RCV), to determine statistically significant changes in serial results (15). A combination of the use of a common reference interval with harmonized data for routine clinical biochemistry and observed changes versus baseline values (using changes beyond RCV) as alerts seems to be the most optimal set of tools to assess safety in clinical studies. REFERENCES 1. Chuang-Stein C. Laboratory data in clinical trials: a statistician´s perspective. Drug Inf J. 1998; 19:167–177. 2. Thompson WL, et al. Routine laboratory tests in clinical trials: interpretation of results. J Clin Res Drug Dev. 1987;1:95–1 19. 3. Ruvuna F, et al. Generalized lab norms for standardizing data from multiple laboratories. Drug Inf J. 2003;37:61–79. 4. NCCLS C29-A. How to define and determine reference intervals in the clinical laboratory. Approved guideline, 2nd ed. Villanova, PA: National Committee for Clinical Laboratory Standards; 2000. 5. Jones GRD, et al. The case for common reference intervals. Clin Biochem Rev. 2004;25:99–104. 6. Leslie WD, Greenberg ID. Reference range determination: the problem of small sample sizes. J Nuclear Med. 1991;32:2306–2310. 7. Sunderman F, et al. Current concepts of “normal values, “reference values” and “discrimination values.” Clin Chem. 1975;21:1873–1877. 8. Fraser CG. Inherent biological variation and reference values. Clin Chem Lab Med. 2004;42: 758–764. 9. Baadenhuijsen H, et al. A model for harmonization of routine chemistry results between clinical laboratories. Ann Clin Biochem. 2000;37: 330–337. 10. Myara A, et al. Harmonization of liver enzyme results: calibration for aminotransferases and gamma glutamyltransferase. J Hepatol. 2004;41: 499–504. 1 Scholten R. Evaluation of hemoglobin QC at a 1. number of regional laboratories. Internal study on file at Covance, Indianapolis. 2007. 12. Barry PL, Westgard JO. Method validation: refer- Drug Information Journal
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.