21-hydroxylase autoantibodies (21OHAb) are the gold standard immune marker to identify patients with clinical or subclinical autoimmune Addison's disease (MD). No assessment of interlaboratory concordance has been made for 21OHAb measurement. Serum samples from 267 patients with primary adrenal insufficiency and from 83 healthy control subjects were distributed to four independent laboratories that determined presence and titer of 21OHAb, by using radiobinding assays with either in vitro translated S-35-radiolabelled or I-125-radiolabelled autoantigen. Cohen's kappa of inter-rater agreement ranged from 0.857 to 0.983, showing a very good concordance of the positive/negative score among the four laboratories. Passing-Bablok regression showed a good agreement of 21OHAb titers arranged by ranks, but important discrepancies emerged at the Bland-Altman plot, as the repeatability coefficient was much higher than the laboratory cut-offs, which indicates that results from different laboratories cannot be used interchangeably. A standardization international program for 21OHAb measurement is strongly needed.
Measuring adrenal autoantibody response: interlaboratory concordance in the first international serum exchange for the determination of 21-hydroxylase autoantibodies
FALORNI, Alberto;BINI, Vittorio;
2011
Abstract
21-hydroxylase autoantibodies (21OHAb) are the gold standard immune marker to identify patients with clinical or subclinical autoimmune Addison's disease (MD). No assessment of interlaboratory concordance has been made for 21OHAb measurement. Serum samples from 267 patients with primary adrenal insufficiency and from 83 healthy control subjects were distributed to four independent laboratories that determined presence and titer of 21OHAb, by using radiobinding assays with either in vitro translated S-35-radiolabelled or I-125-radiolabelled autoantigen. Cohen's kappa of inter-rater agreement ranged from 0.857 to 0.983, showing a very good concordance of the positive/negative score among the four laboratories. Passing-Bablok regression showed a good agreement of 21OHAb titers arranged by ranks, but important discrepancies emerged at the Bland-Altman plot, as the repeatability coefficient was much higher than the laboratory cut-offs, which indicates that results from different laboratories cannot be used interchangeably. A standardization international program for 21OHAb measurement is strongly needed.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
