Upsala J Med Sci 99: 357-362, 1994 7.5 Common Reference Intervals for Plasma Proteins in the Nordic Countries Ole Blaabjergl, Per Hyltoft Petersenl, Mogens Blom2, Kerttu Irjala3, Adam Uldal14, Hanne Cry5, Karla Mattila3, Irma Matinlauri3, Erik Lund6, Jens Rahbek Norgaardl. 1. Department of Clinical Chemistry, Odense University Hospital, DK-5000 Odense C, Denmark. 2. Department of Clinical Chemistry, Hjgrring Sygehus, DK-9800 H j ~ r i n g , Denmark. 3, Department of Clinical Chemistry, T u r k u University Hospital, SF 20520 Turku, Finland. 4 . Department of Clinical Chemistry, KAS Herleu, DK-2730 Herleu, Denmark. 5. Medi-lab, Adelgade 5-7, DK-1304 Copenhagen, Denmark. 6. Department of Clinical Chemistry, Vejle Sygehus, DK-7100 Vejle, Denmark. It is a philosophic question whether it is possible to establish common reference intervals valid for different ethnic groups and geographical areas. From a practical viewpoint, however, reference intervals are in current use and seem to be the best tool for a general validation of the first results from persons consulting the health care system - when clinical strategies with clear interpretation of results are missing. The reference individuals should be selected and the measurements and calculations should be performed according to IFCC (1, 3) or equivalent (cf. section 7.1). A n important point, however, is the decisions about dwiding and combining different reference intervals as investigated by Harris and Boyd (2). According to them the problem is mainly a statistical problem, but as discussed in chapters 7.1, 7.2 and 7.3, it is also a question of judgements based on biology. Presumptions for Sharing Common Reference Intervals Three main conditions have to be fulfilled for establishing common reference intervals: I H The analytical quality must be common The biology must be common, and The interpretation of data must be common. 357 The analytical quality must be common In principle it is sufficient just t o have the same analytical quality in all the laboratories involved, but in practice this can only be obtained by using either the same method (calibrator, reagents, equipment, instructions etc.) or by using specific methods with first class calibrators with values traceable t o the highest level of trueness. Only the last solution is of interest for plasma proteins. The IFCC/CAP/ BCR 470 reference preparation for plasma proteins makes it possible to obtain the needed standardization. The Nordic Calibrator has concentration values traceable to this reference preparation (cf. chapter 5). The main problem of plasma protein analyses, the unspecific reactions from turbid samples for nephelometric methods can be solved by ultracentrifugation of the samples (cf. chapter 6). So, the analytical prerequisites are available for establishing common reference intervals. The biology must be common We have investigated two ethnic groups living in different geographical areas in the Nordic countries. We d d only find a dfference for S-Haptoglobin of such an importance that we decided to specify different reference intervals in Finland and Denmark for this protein, probably due to differences in genetical subtypes of this protein. The results for immunoglobulins indicated variations, which may be of importance, but the tendencies between the two sexes were conflicting and could not justify any splitting up in Finnish and Danish reference intervals, as the groups used for the comparison were too small in size and only related t o the age group between 30 and 40 years of age. The question about the immunoglobulins may be clarified by Swedish investigations where no increase in S-IgA with increasing age was found (personal communication). Thus, differences in biology may be disclosed by large scale investigations designed according to more specific questions. The interpretation of data must be common We have not used the IFCC recommendations for estimation of the reference intervals as we wanted to use the probit display in order to validate possible combinings of reference intervals for different groups (cf. sections 7.1 and 7.2). The results, however, are valid as the distributions are close to log-Gaussian (sections 7.2, 7.3, and 7.4). The comparison between Finnish and Danish showed that in the Nordic countries it could be possible to establish common reference intervals for the seven proteins, with S-Haptoglobin and S-IgA as exceptions - and with the methodological exceptions for S- al-Antitrypsin (cf. section 7.3). 358 Reference Intervals for Nine Plasma Proteins Group M (all) W ( + E ) M (>50) W i + E ) W ( > 50) W ( > 50) Table 7.5.1 Reference interval 0.26-0.45 36.6-48.2 P r o t e i n Prealbumin Transthyretin M ( < 5 0 ) W (-E) Albumin * 39.6-51.1 Orosomucoid al-Acid Glycoprotein a,-Antitrypsin ** al-Trypsin Inhibitor Haptoglobin *** Transferrin If@ **** Immunoglobulin A IgG Immunoglobulin G IgM Immunoglobulin M Plasma Proteins (conc. in g/L) (96 %-intervals) I M (all) M ( > 50) D 0.47-2.05 W b 5 0 ) F 0.32-1.90 M (all) W ( > 5 0 ) 6.1-14.9 W (-E) 0.23-0.39 ---t--- W (-El I 0.45-1.08 1 W (-E) M ( < 5 0 ) W (-El W ( + E ) t W ( + E l 6.9- 15.7 0.70-3.65 W (-E) W (-E) W ( + E ) 1 0.55-2.30 I Type MZ (0.60-0.99) --t- I M ( a l l ) = allmen, M ( > 50) = M ( <50) = W ( > 50) = W (-E) = W ( + E) = * ** men over 50 years, men under 50 years, women over 50 years, women under 50 years, not using estrogens, women under 50 years, using estrogens, The reference intervals for S-Albumin include an analytic CV, - 3.4 %. Reference intervals for Type MM and for MS and MZ (without estrogen The results for S-a,-Antitrypsin are only valid when antisera from DAK are used. See text and chapters 5 and 6. D = Danes, F = Finns. Slightly lower S-Haptoglobin-values in the Finnish population, approx. 0.15 g/L. PP). *** **** Non-parametric estimates. All individuals sitting at least 15 min before blood sampling in arm-vein. 359 Use of different units (grams or moles) together with minor differences in presentation may, however, give a heterogeneous impression. So, even if the same material with the same method for estimation of the reference intervals have been used the list will be different in each of the Nordic countries. This situation was not foreseen at any time during the projects, and it was a great surprise to us. The table with reference intervals for the nine plasma proteins are the results estimated by the project group - with values traceable to CRM 470. Discussion There are several week points in the reference intervals produced. Analytical problems Concerning S-al-Antitrypsin a problem with the transfer of values from CRM 470 to the Nordic calibrator exists (cf. chapter 5). Therefore, at present we can only guarantee the reference intervals for this protein when DAKO antisera are used. For the other eight proteins there should be no problems as long as the patient samples are not turbid. Turbid samples for nephelometry should be ultracentrifuged. Estimution of the reference intervals The division of reference intervals into two groups for the eight proteins (three for S-Orosomucoid) is more or less obvious. In more questionable cases it has been our decision and we are responsible for the conclusion in each case. In all the cases with convincing log-Gaussian distributions the division seems to be reasonable, but for S- Haptoglobin and S-IgA the non-log-Gaussian distributions point to problems indicating that the &vision is not optimal. A r e the reference intervals common ? The question whether the reference intervals are common for the Nor&c countries - except from S-Haptoglobin and S-IgA can only be answered by a new and extended investigation with a special design for the clarification of the weaker points in this work. Are hospital employees and their relutives representative ? This question cannot be answered by this investigation, but it should be examined in another project. 360 General discussion The present situation where each laboratory has its own reference interval for each quantity is not justified by the analytical methods in use. Even with such a performance due t o the analytical method, the procedure cannot be characterized as professional, as there are no biological reasons justifjmg different reference intervals except from ethnic differences and environmental conditions - and this has not been taken into consideration when laboratory-individual reference intervals were established. There are good reasons for establishing common reference intervals in the Nordic countries - and even with the incomplete attempt described here for the nine plasma proteins - the use of common reference intervals is the only way to strive for. The analytical possibilities based on common calibration and specific methods are the logical consequence for the plasma proteins. It is in any case better than to maintain the situation of laboratory-individual reference intervals. Acknowledgements We want to thank all the persons who have contributed to this work, volunteers who gave blood to the investigation and the 20 Danish laboratories involved in collecting the blood samples and doing extra analytical work. We are specially in debt of gratitude to Inger Nwgaard, who performed the protein determinations, and the agarose electrophoreses for estimation of M-Components, and to h e Richter, who performed the registrations and calculations of the many results. Further we are grateful to Soren Blirup and Per Just Svendsen, who performed the measurements of S-CRP and for the antisera used in all the determinations delivered by DAKO . References 1. Alstrom T, Grasbeck R, Lindblad B, Solberg HE, Winkel P, Viinikka L. Establishing reference values from adults: recommendation on procedures for the preparation of individuals, collection of blood, and handling and storage of specimens. Seand J Clin Lab Invest 1993;53649-52. Harris EK, Boyd JC. On Dividing Reference Data into Subgroups to Produce Separate Reference Ranges. Clin Chem 1990;36:265-70. Solberg HE. Approved Recommendations (1987) on the Theory of Reference Values. Part 5. J Clin Chem Clin Biochem 1987:25:645-56. 2. 3. 36 1