Evin Ademoglu, Okhan Akin, Ali Riza Sisman, Nurzen Sezgin, Mustafa Orkmez, Sehavet Tezcan, Ali Okuyucu, Melahat Dirican, Abdulkerim Bedir, Turkan Nogay, Doğan Yücel, Pinar Akan, Ayse Binnur Erbagci, GulOzlem Tuncer, Mehmet Gurbilek, Ebubekir Bakan, Ozgur Kadicesme, Murat Can, Selda Erdinc, Canan Çoker, Onur Bahceci, Fatma Taneli, Diler Aslan, Nezaket Eren, Mehmet Aykus, Harun Polat, Nuriye Mete, Elif Degirmen, Damla Kayaalp, Kübra Doğan, Sebahat Ozdem, Yeliz Gunes, Necip Ilhan, Reo Kawano, Dilek Sadak Atali, Osman Evliyaoglu, Orhan Deger, Cigdem Damla Cetinkaya, Husamettin Vatansev, Muhittin Serdar, Bahadir Ozturk, Halide Akbas, Sevda Unalli Ozmen, Yesim Ozarda, Kiyoshi Ichihara, Meltem Demir, Gultekin Yucel, Aysel Kiyici, Tevfik Noyan, Gürbüz Polat, Ismail Kurt, Hulya Aybek, Ayşem Kaya, Ozlem Demirpence, Suret Agac, Asli Pinar, Zeki Ari, Zonguldak Bülent Ecevit Üniversitesi, and Ondokuz Mayıs Üniversitesi
Background: A nationwide multicenter study was organized to establish reference intervals (RIs) in the Turkish population for 25 commonly tested biochemical analytes and to explore sources of variation in reference values, including regionality. Methods: Blood samples were collected nationwide in 28 laboratories from the seven regions (?400 samples/region, 3066 in all). The sera were collectively analyzed in Uludag University in Bursa using Abbott reagents and analyzer. Reference materials were used for standardization of test results. After secondary exclusion using the latent abnormal values exclusion method, RIs were derived by a parametric method employing the modified Box-Cox formula and compared with the RIs by the non-parametric method. Three-level nested ANOVA was used to evaluate variations among sexes, ages and regions. Associations between test results and age, body mass index (BMI) and region were determined by multiple regression analysis (MRA). Results: By ANOVA, differences of reference values among seven regions were significant in none of the 25 analytes. Significant sex-related and age-related differences were observed for 10 and seven analytes, respectively. MRA revealed BMI-related changes in results for uric acid, glucose, triglycerides, high-density lipoprotein (HDL)-cholesterol, alanine aminotransferase, and ?-glutamyltransferase. Their RIs were thus derived by applying stricter criteria excluding individuals with BMI >28 kg/m2. Ranges of RIs by non-parametric method were wider than those by parametric method especially for those analytes affected by BMI. Conclusions: With the lack of regional differences and the well-standardized status of test results, the RIs derived from this nationwide study can be used for the entire Turkish population. © by De Gruyter 2014., Applied Scientific Research Fund: :24256003:2012-2014 Japan Society for the Promotion of Science, Acknowledgments: This study was supported by Research Fund of Uludag Universty (UAP(T)-2011/48), IFCC C-RIDL, Scientific Research Fund (No:24256003:2012-2014) provided by the Japan Society for the Promotion of Science, Abbott Diagnostics and Becton Dickinson. We had invaluable input from the IFCC C-RIDL and the Turkish National Biochemical Association. We are very grateful to Uludag University Central Laboratory, We are thankful to David Ambruster from Abbott Diagnostics for his kind cooperation to realize this study. The final manuscript was edited with generous support by Prof. A. Myron Johnson.