Quantification of CEA by the microparticle capture method on the ... 95% of sera from a group of blood donors gave values
tions of reference limits. In: Gr#{225}sbeck it, nous IgG, AhistrOm T, eds. Reference values in laboof RF to ratory medicine. Norwich, CT: Wiley, 1981: with the
193-205.
10. Fleisher M, Niseelbaum JS, Loftin L, Smith C, Schwarts MK. Roche RIA and Abbott EIA carcinoembryonic antigen assays compared. Clin Chem 1984;30:200-5. 11. Klee GG, Dodge LA, Reynoso G. Discrepancies in carcinoembryonic antigen measurements: surveyand control values vs values for patients. Clin Chem 1987;33: 563-6. 12. Delfert D, GeorgeS, Rojewski M, et al. Quantification of CEA by the microparticle capture method on the automated IMx instrument [Abstract]. Clin Chem 1990;36: 1095.
13. BOrmer OP, Thrane-Steen
K. Epitope
of six immunoassays for carcinoembryonic antigen. Tumor Biol group specificity
1991;12:9-15. 14. B#{246}rmer OP. Major disagreement be-
tween immunoassays of carcinoembryonic antigenmay be causedby nonspecific crossreacting antigen 2 (NCA-2). Clin Chem 1991;37:1736-9. Ursula
Turpeinen’
Caj Haglund2 Peter Roberts2 UIf-H*kan Stenman’ 1Dept. Obstet. and Gynecol., Lab.,
2Foh Dept. Surgery Helsinki Univ. Central Hosp. Haartmaninkatu 2,00290 Helsinki, Finland
Healthy IndivIduals RheumatoId Factor
Seroposltive
for
To the Editor: Several authors have described frequencies of seropositive results for rheumatoid factor (RF) in healthy individuals without rheumatoid arthritis (1,2). In a recent paper in this journal, Ailus et aL (3) showthis prevalence in a subpopulation of pregnant women. However, there is no consensus on the extent of this clinical fa]se-positivity (true positivity being given by subjects with rheumatoid arthritis diagnosed according to criteria of The American College of Rheumatology). These rates are probably mainly dependent on the methodology used, as Ailus et al. show. We would like to mention two ideas. Many commercial kits for RF measurement have the drawback of requiring heat inactivation of sera at 56 #{176}C for 30 mm to denature the complement componentClq, which causes a positive interference in some RF assays. Nevertheless, heat inactivation produces erroneous results by leading to the aggregation of endoge-
which inhibits the binding the aggregated IgO added reagents. Clq interference
can easily be avoided by including a chemical inhibitor of Clq (polyvinyl
suiphonate, sodium salt) in the reaction medium (4). Second, using two methods that do not require sample pretreatment (5), we recently
confirmed that some RF-posi-
tive aera become negative
after
heat
treatment, as determined with a nephelometric or a turbidimetric RF test. These results indicate that methods requiring heat inactivation of sera may present a high rate of false-negative results, especially for RF concentrations in the rangeof 20-45 kItJ/L, and should not be usedfor RF quantification. Establishing RF reference values is very difficult because of the poor analytical sensitivity of the various available methods and the extremely low RF values in healthy individuals. In many RF assays,the upper “normal” limit is not in accordance with recommendations of the International Federation of Clinical Chemistry or simply is not stated (6-8). We found that 95% of sera from a group of blood donors gave values
