Search

Most hospital laboratories estimate the concentration of total circulating testosterone using a non-extraction method on an automated multi-channel immunoassay analyser supplied by a small number of multi-national diagnostic companies. Although these platforms offer advantages of quick turnaround times, small volume sampling and random access analysis, proficiency testing schemes suggest the quality of results produced remains similar to that of the early manual radioimmunoassay. An estimate of the bioavailable, non-sex hormone binding globulin (SHBG) bound fraction of circulating testosterone, be that the free or the free plus albumin-bound, may be a better index of gonadal status than total testosterone alone, especially when a borderline hypogonadal level of total testosterone is found, and may avoid misclassification of hypogonadal or eugonadal men. Free or bioavailable testosterone may be calculated or measured. The free androgen index may not give a true reflection of androgen status in men. In the interpretation of serum testosterone concentrations with results >40 nmol/L, the possibility of exogenous administration or abuse needs to be considered. The marked diurnal rhythm in total testosterone should also be taken into account. There may be a diminution of testosterone secretion with advancing age, but the great majority of older men have a circulating total testosterone concentration well within the accepted reference intervals established for younger men. As testosterone concentration may fluctuate markedly both seasonally and from day to day, it may be judicious to measure levels on more than one occasion. Provided that estimates of serum testosterone are unequivocally eugonadal (12.5-40 nmol/L) or hypogonadal (<7.0 nmol/L), results produced by routine automated immunoassays will in all probability give a satisfactory assessment of androgen status in men.Routine biochemical assessment of gonadal function in men should include measurement of early morning luteinizing hormone, follicle stimulating hormone, prolactin and SHBG together with total testosterone, and if necessary some estimate of bioactive testosterone.

Concentrations of 14 commonly-requested plasma hormones were measured in octuplicate in each of six subjects to determine their stability when unseparated from red cells for periods up to 1 week. Most of the analytes were stable when stored in this way and although statistically significant changes were recorded, in the great majority of cases the changes seen would have no bearing on the clinical interpretation of the result. In the light of these findings, we would confidently report results of analyses for these hormones in plasma that had remained in contact with red cells at ambient temperature for long periods of time.

This paper summarises the views of the authors on the provision of a prolactin assay service. We discuss the pathophysiology of prolactin secretion and the clinical indications that arise from that. We cover the rather complex issue of the definition of normal and elevated prolactin levels. From these considerations, certain guidelines on the analytical performance of prolactin assays and their provision in a clinical biochemistry service are given. The extent to which currently available methods and performance as revealed by the UK External Quality Assessment Scheme (EQAS) match these guidelines are described and certain conclusions are reached. Finally, probable future developments are briefly discussed. The main conclusions and recommendations are as follows: Reagents of appropriate quality are available to enable prolactin immunoassays to be provided in UK clinical biochemistry laboratories. These are provided either separately or in the form of kits from both commercial and NHS sources. There is no requirement for individual laboratories to undertake their own antiserum production or prolactin iodination. Acceptable performance (as defined using internal QC procedures and the UK EQAS) is achievable using these reagents/kits, although one commercial kit shows a consistent marked negative bias. Reference ranges, including 'normal ranges', show considerable between-centre variability. Many centres have not established their own ranges, even those using in-house methods. Reference ranges for use in clinical biochemistry laboratories are proposed in this report. Some general guidance on the provision of a prolactin service is given, although this does not differ in principle from that appropriate for other peptide hormone analytes. There is no evidence that centres with small workloads perform any worse than average, although it may be more cost-efficient for such centres to send the samples elsewhere. As with other peptide analytes, non-isotopic immunometric methodology is likely to replace current radioimmunoassay methods in the near future.