HEALTHY YOUNG MAN WITH ”LOWISH FREE T”, NEEDING TREATMENT??
QUESTION- I am an endocrinologist working in a small hospital in Somerset, UK. We are coming up against a testosterone related clinical issue as follows –
We are seeing a number of younger men who are requesting investigation for hypogonadism and coming up with low-ish testosterone levels. The issue relates to how to interpret the results in the context of our local assay. We use the Beckman Access Testosterone Assay that quotes a normal range (from the manufacturer and not age related) of 6.07-27nmol/l.
The case that I am struggling with at present illustrates the issue: 29 year old referred because of fertility issues and is requesting a testosterone prescription. He has non-specific symptoms such as tiredness, but no sweats and only occasional ED(possibly a slightly reduced libido).
Clinically he has no features of hypogonadism and has normal sized testes (25ml bilaterally).
His blood results are:
Total testosterones (9am samples) 6.7nmol/l, 6.9nmol/l and 5.1nmol/l
SHBG 10nmol/l (NR in the lab we use 10-50nml/l)
Calculated free testosterones 0.201nmol/l, 0.208nmol/l and 0.152nmol/l (using http://www.issam.ch/freetesto.htm )
The relevant pituitary function results are: LH 7.4u/l, FSH 5.2u/l, PRL 178miu/l
Between appointments his partner has fallen pregnant(!). His semen analysis was 25- 30mill/ml but with 95% abnormal morphology.
My questions are:
1. Generally - the combined guidelines from the ASA, EAA, EAU, ISA, and ISSAM recommend obtaining free testosterone levels in men with total testosterone 8–12 nmol/L, with a suggested threshold of 225.4 pmol/L established as the lower range of normal. However, given our assay has a quoted normal range down to 6.07nmol/l, which is likely to be a reflection of a mixed age population of normals, we will not infrequently encounter men with low calculated free testosterone - are they really testosterone deficient?
- For the case above – Is the fact that he has fathered a child enough biological evidence to say that he does not need testosterone, or does the abnormal sperm morphology with the other softer features mean that a trial of therapy with close monitoring of testosterone would be the way forward?
I look forward to hearing from you. Dr Alex Bickerton BM FRCP DPhil (Oxon),Yeovil District Hospital NHS Foundation Trust
RESPONSE--A 29-year-old man is referred because of fertility issues and was requesting a testosterone prescription. He has non-specific symptoms including tiredness, slightly reduced libido and occasional erectile dysfunction. Clinically he has no significant co-morbidities or features of hypogonadism. He is not obese and has 25ml testes bilaterally.
His morning serum hormone results are:
- Total testosterone (9am samples) 6.7, 6.9 and 5.1nmol/l (1.93, 1.99, 1.48ng/ml)
[Beckman Access Testosterone Assay, ref range 6.1-27nmol/l. = 1.76- 7.79 ng/ml]
- SHBG 10nmol/l (Ref range 10-50nmol/l)
- Albumin 46g/l
- Calculated free testosterones 0.201, 0.208 and 0.152nmol/l (calculated using http://www.issam.ch/freetesto.htm )
Selected pituitary function tests are normal LH 7.4 IU/l, FSH 5.2 IU/L, PRL 178 IU/L
Semen analysis: Sperm concentration was 30 million/ml with 5% normal morphology.
Between appointments his partner conceived has naturally!
Generally - the combined guidelines from the ASA, EAA, EAU, ISA, and ISSAM recommend obtaining free testosterone levels in men with total testosterone 8–12 nmol/L, with a suggested threshold of 225.4 pmol/L established as the lower range of normal. However, given our assay has a quoted normal range down to 6.07nmol/l, which is likely to be a reflection of a mixed age population of normal, we will not infrequently encounter men with low calculated free testosterone - are they really testosterone deficient?
“Total T is the key’: All guidelines direct that the diagnosis of androgen deficiency (AD) be based on a synthesis of clinical features and confirmatory biochemistry, and rely on total testosterone (TT) as the key measure. One must take into account all these aspects and cannot use a parameter, such as free T, to exercise a ‘casting vote’. This measure is not analogous to free thyroid hormones and its place in clinical decision making is limited (underscoring the vague references to its use in the guidelines) and some would argue, non-existent. It can be well argued that the free T measure does not provide a better measure of androgen status than TT. Nonetheless its use has become widespread and it is now often provided without specific request and without validates reference ranges – this is extremely problematic as discussed below.
Consideration of free T measures is suggested when changes in SHBG are marked and in turn affect total testosterone levels and their interpretation. Mild elevation in SHBG levels is seen with ageing, but marked elevation is seen hyperthyroidism, liver disease and anti-epileptic therapies that are accompanied by elevated TT without evidence of androgen excess. Conversely SHBG levels are low in obesity, insulin resistance, and exposure to exogenous androgens, in which setting serum TT levels fall below the reference range, especially with normal serum LH and FSH, but do not confirm a diagnosis of androgen deficiency (1). An intrinsically low normal SHBG in this case probably accounts for the low range TT level but does not inform the diagnosis of AD and thus treatment.
There are different approaches to assessing free T with direct measure by equilibrium dialysis (EQ) being the most scientifically valid yet rarely available in practice. Estimated (calculated) free T measures are based on various formulae that use TT, SHBG and albumen levels along with assumptions about binding affinities. Close examination shows that the correlate reasonably well with EQ free T yet generate significantly different values using the same primary data (2).
In all formulae, the TT estimate has the major effect on the calculated free T readout; this raises the next problem, of how well are TT assays standardised? The news here is not good; there is considerable variability in TT levels between immunoassay platforms when applied to identical samples. As a demonstration, we studied reproductively normal young men (normal testes and semen analysis), the reference range for TT using mass spectrometry was 10.4-30.1 nmol/L but the lower limit of the reference ranges across 7 commercial immunoassay methods ranged from 7.3 to 12.1nM! (3) There is inherent variability between immunoassay platforms (3,4) yet they dominate commercial laboratories due to their high throughput and low cost. There is a global movement toward quantitative approaches, specifically mass spectroscopy [MS] and these are becoming more available, despite the high set up costs. Given that calculated free T is highly dependent upon the TT values, any attempt at a consensus lower limit figure, [such as 225pmol/L] is a ‘rubbery’ figure indeed!
Perhaps it is also worth reflecting on TT reference intervals: what is a healthy normal TT? Currently laboratories report a ‘one size fits all’ interval for normal men, irrespective of age, adiposity and co-morbidities. While this patient is young, issue often arise in middle aged and older men. It is becoming clear that the age-related decline in serum testosterone is primarily a result of the accumulation of co-morbidities with age, especially obesity which is the major determinant of serum T within population. No decline was seen in serum TT of healthy aging men (5) and there is little objective support for an intrinsic decline in the HPT axis function in male ageing (the so called ‘andropause’). The European Male Ageing Study (EMAS) shows this impact if obesity: across middle aged and older men, obese men (BMI > 30) have serum T level ~ 6 nM lower than their lean peers (6). One encounters many obese and/or diabetic men with TT below the reference range; a low SHBG may result in a normal, or at least somewhat higher, calculated free T. There is no way of knowing but the inclusion of older men with co-morbidities may account for the low end reference value of 6.07nmol/L in the questioner’s lab although a recent study very healthy men aged 70-89 years, provided a reference range using mass spectrometry of 6.4-25.7 nmol/L (7).
It is clear that in are many men presenting for evaluation of possible androgen deficiency, TT levels will be below that provided in the kit manufacturer package insert emphasising the need for clinical judgment and not simply to rely on a numerical threshold.
In this particular case, there seems no evidence of androgen deficiency. He is clinically well virilised, demonstrably fertile and has normal testes. Objective tissue markers of androgen action include haemoglogin and bone density would likely confirm eugonadism. His SHBG falls in the lower end of normal and appears not to reflect any pathological process but substantially explains his low range TT. The calculated free TT value of ~200pmol/L which is around the lower range of normal and does not add to the clinical management. His symptoms may reflect an entirely different process.
- Is the fact that he has fathered a child enough biological evidence to say that he does not need testosterone, or does the abnormal sperm morphology with the other softer features mean that a trial of therapy with close monitoring of testosterone would be the way forward?
This question has two components:
2.1 No, infertile men with subnormal semen quality may have normal or low TT levels: an elevated serum LH being indicative of primary AD due to testicular failure. Conversely, many men with poor semen quality may conceive naturally; around 30% of couples get pregnant with sperm concentration in the 1-5 million/ml range. Thus one cannot conflate the issues of androgen deficiency (Leydig cell function) and fertility (spermatogenesis). Although making it much less likely, the fact that he caused a natural conception does not mean he could not have been androgen deficient. But in this case, he has normal semen, normal sized testes, mid normal gonadotropins and low-normal TT, SHBG and free T.
2.1 I don’t exactly follow this proposition. I do not see any indication for a ‘trial’ of testosterone therapy as this man does not have demonstrated testicular or pituitary problems and is clinically eugonadal. Also, testosterone is a potent contraceptive and is contraindicated in men seeking fertility. One does not know the outcome of the current pregnancy nor whether subsequent ones are planned. Exogenous testosterone will suppress LH, FSH and sperm output and testis size will fall. His sperm morphology is of no importance and ‘softer testes’ (noting they are 25 ml in volume) is a subjective parameter which would not direct management. Thus I would re assure him re his male hormonal status and see no need for treatment or ongoing review in this regard.
Prof Robert McLachlan 1/5/2018
1 Yeap BB, Grossmann M, McLachlan RI et al Endocrine Society of Australia position statement on male hypogonadism (part 1): assessment and indications for testosterone therapy. Med J Aust. 2016 ;205:173-8.
2 Ly LP, Sartorius G, Hull L, et al. Accuracy of calculated free testosterone formulae in men. Clin Endocrinol 2010; 73(3): 382-388.
3 Sikaris K, McLachlan RI, Kazlauskas R, et al. Reproductive hormone reference intervals for healthy fertile young men: evaluation of automated platform assays. J Clin Endocrinol Metab 2005; 90(11): 5928-5936.
4 Taieb J, Mathian B, Millot F, et al. Testosterone measured by 10 immunoassays and by isotope-dilution gas chromatography-mass spectrometry in sera from 116 men, women, and children. Clin Chem 2003; 49(8): 1381-1395.
5 Sartorius G, Spasevska S, Idan A, et al. Serum testosterone, dihydrotestosterone and estradiol concentrations in older men self-reporting very good health: the healthy man study. Clin Endocrinol 2012; 77(5): 755-763.
7 Yeap BB, Alfonso H, Chubb SA, et al. Reference ranges and determinants of testosterone, dihydrotestosterone, and estradiol levels measured using liquid chromatography-tandem mass spectrometry in a population-based cohort of older men. J Clin Endocrinol Metab 2012; 97(11): 4030-4039.