Abstract
For T4, radioimmunoassay (RIA) methods gave values close to the all-method mean. Competitive protein binding (CPB) kits gave values significantly higher (by 12%) in one and lower (by 10%) in another, probably due to the extraction of interfering substances such as non-esterified fatty acids or thyroid hormone binding proteins. The range of between-batch precision of individual laboratories was 10·8–47·5% (mean 17·8%) over the clinically relevant range. It was particularly poor at levels below the lower limit of the normal range.
For T3 RIA, kit methods gave significantly higher values than independent methods although recovery of exogenous T3 was approximately quantitative with both. Cross reaction of T4 in T3 assays was insignificant. The range of between-batch precision for individual laboratories was 11·4–35·5% (mean 21·9%), and was poorest at levels of 1 nmol/l or less.
For digoxin RIA, Lanoxitest γ gave significantly higher values than other methods, and over-recovered exogenous digoxin. The range of between-batch precision was 13·8–35·2% (mean 21·6%), and was poorest at levels of 1 nmol/l or less.
The distribution of liquid human serum specimens at ambient temperature was satisfactory for quality control of T4, T3, and digoxin RIA but not for T4 by the Thyopac 4 method. Ox serum was unsuitable for certain T4 RIA methods.