Proficiency in the detection and titration of blood group antibodies: results of an interlaboratory trial.

Twelve numbered samples of serum were distributed to 82 Canadian blood transfusion laboratories. The presence or absence of blood group antibodies in these samples was correctly reported by less than 50% of participants; inability to demonstrate the presence of weak anti-D or anti-c was responsible for most of the failures. On the other hand, about 85% of participants reported titres of anti-D and anti-Fy(a) within an acceptable range. Two proposals are made that should help laboratories achieve a high and uniform level of proficiency in antibody detection and titration.