Guest Essay

Dr. Pål Rustad explains how more than 100 laboratories in the Nordic countries collaborated to produce common reference intervals for 25 quantities in serum/plasma and 8 hematology quantities.

Common reference intervals for 25 quantities in serum/plasma and 8 hematology quantities in the Nordic countries

Documentation published in a special issue of Scandinavian Journal of Clinical and Laboratory Investigation (SJCLI), Clinical Chemistry and Laboratory Medicine special issue and on the NORIP Project website

Fürst Medical Laboratory, Søren Bulls vei 25, NO-1051 Oslo, Norway

Special topics

• Routine measurement systems used to establish reference values by correcting laboratory bias detected by using a commutable reference material.
• Reference intervals established using methods compatible with the new IFCC recommended methods at 37 °C for lactate dehydrogenase, alkaline phosphatase, amylase and pancreatic amylase.
• Common reference interval for creatinine independent of method used; Jaffé -, enzymatic - or dry chemistry (Vitros, Ortho).

Background

In all Nordic countries the need for traceable reference intervals in clinical chemistry grew through the last decade of the last century. The origin of many of the reference intervals used in laboratories is obscure. Using commutable normal serum in external quality assurance programs it was noticed, and especially in Norway documented, that the reference intervals used within the country for the same quantity in the same age and gender groups varied more than the corresponding analytical deviations could account for. It was believed that the populations in the Nordic countries were too homogenous to make room for a biological explanation. Therefore with the Norwegians as the driving force and supported by the Nordic Society of Clinical Chemistry (NFKK) a Nordic project was launched on March the 27th 1998 in Oslo based on a decentralized design from Denmark. A subproject for routine hematology with origin in Finland was added later (paper 9). The project was performed mainly during 2000 and 2001 in the five Nordic countries according to a common protocol.

Concept

Each of the 102 participating Nordic laboratories selected at least 25 healthy individuals aged 18 years or older. The individuals were evenly distributed in specified gender and age groups, fulfilled specified health criteria and answered questions relevant for possible further partitioning of the reference values. The common protocol for preparation of the individuals and the blood collection corresponded to the handling of outpatients.

The project aimed at producing biological reference intervals with high analytical trueness. However the local routine methods of the participating laboratories were used. Therefore aliquots of a commutable liquid frozen pool of unmodified normal donor serum with high quality target values with uncertainties (controls CAL or NFKK Reference serum X) were measured in 10 replicates in every analytical series of the project and (for non-enzymes) used to proportionally correct the measured values. Also other project controls were used to assess the analytical quality.

Results

The results showed that partitioning of the reference values by country was not necessary. Common Nordic reference intervals are therefore recommended. The special NORIP-issue of SJCLI gives the documentation for these recommendations. However, before routine use of these reference values the laboratories have to demonstrate acceptable trueness when measuring on healthy persons. They can do this by demonstrating acceptable closeness of their results to the target values when measuring on the NFKK reference serum X described in paper 3 and 5.

Paper 1 is what we call the main paper of the project covering all areas and giving the final recommendations. The reference intervals for serum creatinine may be an exception because the reference values measured by the Jaffé method should be recalculated according to measurement of a creatinine-free serum as described in ”Letter to the editor”.

Paper 2 describes the thoughts preceding the project. As mentioned the chosen design of the main project needed a reference material commutable with normal serum samples and with target values of high quality for all measured components at relevant levels. Paper 3 describes the production and stability of the control materials, while paper 4 deals with the establishment of target values both by using reference methods and by transferring values between two commutable materials with routine measurement systems (the Nordic Trueness Project 2002). Paper 5 is a reprint of the first version of the certificate for NFKK Reference Serum X used in the NORIP project.

More details on the reference individuals and the samples are given in paper 6. This paper may be of special interest for users of the bio-bank (see below). The corresponding descriptive detailing of analytical issues is given in paper 7 (see also paper 13). In paper 10 it is shown that deletion of inferior analytical series as judged by other control results relative to those of CAL were of minor importance for the reference intervals.

Establishment of reference intervals for IFCC compatible routine methods run at 37 °C were of special interest, since new recommendations from the IFCC was were released or is about to be released. The data treatment differed from that of other quantities in that only results from measurement systems with documented traceability to the IFCC reference methods were accepted and used as reference values without corrections. This part of the project is therefore described separately in paper 8.

In paper 11 the new partitioning test used in NORIP is compared to the Harris-Boyd method using practical examples from NORIP supporting the chosen non-partition of the reference intervals by country.

The project is not over. We urge our colleagues in the Nordic countries to use the data and the samples from the project. These are stored in the data- and bio-bank described in paper 12.

Content of Special issue of SJCLI (Scand J Clin Lab Invest 2004; 64):

Transnational Biological Reference Intervals

Procedures and Examples from the Nordic Reference Interval Project 2000

1. The Nordic Reference Interval Project 2000:recommended reference intervals for 25 common biochemical properties. Rustad P., Felding P., Franzson L., Kairisto V., A. Lahti, A. Mårtensson, P. Hyltoft Petersen, P. Simonsson, H. Steensland& A. Uldall. Pages 271-284. Abstract
2. Prerequisites for establishing common reference intervals. P. Hyltoft Petersen, P. Rustad. Pages 285-292. Abstract.
3. Minimally processed fresh frozen human reference sera: preparation, testing, and application to international external quality assurance. G. M. Henriksen, M. M. Nørgaard, I. Nørgaard, M. Blom,L. Blou, O. Blaabjerg & A. Uldall. Pages 293-308. Abstract.
4. The Nordic Trueness Project 2002: use of reference measurement procedure values in a general clinical chemistry survey. M. M. Pedersen, U. Örnemark, P. Rustad, H. Steensland, M. Loikkanen, E. Ólafsdóttir, G. M. Henriksen, N. Jørgensen, A. Uldall, G. Nordin & U.-R. Nordberg. Pages 309-320. Abstract.
5. Certificate of analysis: NFKK Reference Serum X: a reprint. M. M. Pedersen, P. Rustad, P. Simonsson. Pages 321-326. No abstract.
6. Reference individuals, blood collection, treatment of samples and descriptive data from the questionnaire in the Nordic Reference Interval Project 2000. P. Felding, P. Rustad, A. Mårtensson, V. Kairisto, L. Franzson, P. Hyltoft Petersen & A. Uldall. Pages 327-342. Abstract.
7. Descriptive analytical data and consequenses for calculation of common reference intervals in the Nordic Reference Interval Project 2000. P. Rustad, P. Felding, A. Lahti, P. Hyltoft Petersen. Pages 343-370. Abstract.
8. Reference intervals for eight enzymes in blood of adult women and men measured in accordance with the IFCC reference system at 37°C - A part of the Nordic Reference Interval Project 2000. J. H. Strømme, P. Rustad, H. Steensland, L. Theodorson, P. Urdal. Pages 371-384. Abstract.
9. A multicentre study of reference intervals for haemoglobin, basic blood cell counts and erythrocyte indices in the adult population of the Nordic countries. G. Nordin, A. Mårtensson, B. Swolin, S. Sandberg, N. J. Christensen, M. Aune, V. Thorsteinsson, L. Franzson, V. Kairisto, E.-R. Savolainen. Pages 385-398. Abstract.
10. Effect of analytical quality on establishing the common reference intervals and their use. P. Rustad, P. Hyltoft Petersen. Pages 399-406. Abstract.
11. Are the common reference intervals truly common? Case studies on stratifying biochemical reference data by countries using two partitioning methods. A. Lahti. Pages 407-430. Abstract.
12. Nordic Reference Interval Project Bio-bank and Database (NOBIDA): a source for future estimation and retrospective evaluation of reference intervals. P. Rustad, P. Simonsson, P. Felding, M. Pedersen . Pages 431-438. Abstract.
13. LETTER TO THE EDITOR: Creatininium reference intervals for corrected methods. A. Mårtensson, P. Rustad, H. Lund, H. Ossowicki. Pages 439-442. No abstract.