Pavilion Health Today
Supporting healthcare professionals to deliver the best patient care

How to diagnose and manage male testosterone deficiency

Testosterone deficiency remains largely under-investigated and untreated in men, possibly due to a misunderstanding of the condition and a misperception of its influence on male health.

Male testosterone deficiency (TD) is a relatively common condition that affects approximately 9% of middle-aged men.1 Among its more recognised symptoms, such as decreased libido, fatigue, and loss of muscle mass, TD is associated with the development of obesity, metabolic syndrome, osteoporosis, and cognitive decline.2 Furthermore, Diabetes UK estimates that approximately 25% of all male patients with type 2 diabetes have lower than normal levels of testosterone.3

Despite its increasing prevalence and ease of diagnosis, TD remains largely under-investigated and untreated in primary care, possibly due to a misunderstanding of the condition and a misperception of its influence on male health. But, as the prevalence of TD continues to rise, it is important that in  the impact of this condition on men’s health is considered and treated appropriately.

To clarify some of the uncertainties associated with male TD, this article offers 10 tips for making an assessment of TD and providing treatment.

1. Don’t forget testosterone as part of the TATT (tired all the time) screen

Testosterone is the main androgen in men. It is essential for the development and maintenance of secondary sexual characteristics, but also has effects on metabolism cognition, cardiovascular function and bone density.4-7 In the brain, it is responsible for stimulating libido, aggression, and aiding cognition, memory and emotions. Up to 7% of GP consultations are related to patient fatigue,8 and its often vague presentation will lead to the majority of affected patients having blood tests. Even though TD commonly presents with fatigue and tiredness, testosterone screening is not included in the investigative algorithm for TATT in GP online,9,10 or even GP Notebook.11 A possible explanation is that testosterone levels are known to decrease with age, so that less importance is placed on testing. Nevertheless, with such a high prevalence of TD, it would seem prudent to include testosterone screening as part of a TATT screen for men over 40 or those of any age presenting with specific TD symptoms.

2. Ageing is not a reason to dismiss low levels of testosterone

Symptoms and signs of late onset TD are varied, and the majority have a multifactorial aetiology. Many of the more non-specific symptoms, such as tiredness, fatigue, or loss of sex drive, are reminiscent of the normal ageing process, and may be found in men with normal testosterone levels.12 This potential confound points to the importance of biochemical assay in confirming low testosterone in those patients with clinical symptoms of TD.

Wu et al13 examined the relationship between age-related decline in testosterone production and the development of comorbidities, and in turn their effect on testosterone levels. They concluded that most of the age-related decline in testosterone production was, in fact, due to the development of co-morbidities, in particular, obesity.

3. Testosterone levels are altered by a range of causes

Many medical comorbidities can affect testosterone levels. Some of the more common ones, in diminishing order of significance, are obesity, type 2 diabetes, hypertension, hyperlipidaemia, osteoporosis, and asthma or chronic obstructive pulmonary disease.14 Other pathological causes of TD include liver disease, obstructive sleep apnoea, significant renal disease, hypothyroidism, and chronic inflammatory conditions.

Conditions such as Cushing syndrome, type 2 diabetes, and cardiovascular disease are associated with obesity and tend to cause a decrease in testosterone levels through a suppressive effect on the hypothalamic-pituitary-gonadal (HPG) axis. In addition, obesity directly reduces sex hormone binding globulin (SHBG), and the presence of increased adipose tissue promotes aromatase activity, which converts testosterone to estradiol.15 TD may be made worse still according to the hypothesis that raised estradiol reduces LH production by suppressing the LH response to GnRH in males.15 The absence of an association with obesity may be why TD is not seen in patients with type 1 diabetes.16

Liver disease causes a direct increase in circulating oestrogen, but cirrhosis can also directly damage the testes and reduce GnRH secretion. Chronic renal failure impairs spermatogenesis, steroidogenesis, and sexual function, largely through negative effects on the HPG axis.15

In addition to chronic diseases, there are also iatrogenic causes of TD. Many commonly used medications can affect testosterone levels through various mechanisms of action. Table 1 highlights some of the more common iatrogenic causes of testosterone suppression and their proposed mechanisms of action.15

Regarding side effects in patients taking TD-promoting medication, treatment should be based on a holistic approach to the patient and their co-morbidities. While some medications may be stopped or changed (for example, switching diuretics), others, such as corticosteroids, are more complex owing to the difficulty in finding alternative treatment options that produce the same desired response. In those patients, testosterone therapy should be considered in conjunction with a reduction, where possible, in the use of testosterone-inhibiting drugs.

On a more positive note, even mild changes in lifestyle can improve testosterone levels. In obese patients, a reduction in 10% bodyweight can increase testosterone levels by up to 4nmol. This effect is probably attributable to a reduction in obesity-induced LH suppression, reduced conversion of testosterone to estradiol, and an increase in SHBG.15

Other lifestyle measures can also raise testosterone levels – or at least minimise its decline. For example, short-term abstinence from sexual activity produces a surge in testosterone at around three weeks,17 but abstinence for more than three months has been shown to reduce testosterone production.18 Although the precise causative mechanism has yet to be confirmed, it has been hypothesised that it is due to an impaired bioactivity of LH found in non-sexually active men.19

Exercise has long been associated with an increase in male testosterone levels, and while the most effective type of exercise is still unclear, recent studies have shown that regular aerobic exercise in overweight individuals produces a significant rise in testosterone levels.20 Conversely, excessive exercise can actually inhibit testosterone production.21

‘Overtraining syndrome’ induces a negative feedback on men’s hypothalamic-pituitary-gonadal axes, with a measurable decline in LH, FSH, and testosterone levels, and an increase in SHBG22 seen in persistent moderate-to-high intensity exercise. This down-regulation of the hypothalamic-pituitary-gonadal access is probably a mechanism for encouraging rest and promoting recovery.

4. Test for it correctly, and don’t necessarily rely on total testosterone

Testosterone production follows a circadian rhythm, with levels peaking between 8am and 11am. The current recommendation is that for accurate assessment, blood tests should be taken before 10:30am and measured by a reliable assay on two or more occasions.23 This protocol has implications for requesting patients’ blood tests, especially for shift workers, who should have their tests done approximately three hours after waking.

Patients found initially to have low testosterone levels should also have serum prolactin, LH, FSH, and SHBG measured on a subsequent retest. There are two reasons for this: determination of serum FSH, LH, and prolactin levels helps exclude secondary hypogonadism; and SHBG levels assist in measuring the ‘true’ levels of testosterone.24 Approximately 60% of serum testosterone is bound to SHBG and a further 38% is loosely bound to albumin and other binding proteins. It is, therefore, only the remaining 2% testosterone that is considered biologically active. As levels of SHBG increase with age, it is sensible to use an online calculator, such as that provided by the Primary Care Testosterone Advisory Group,25 to calculate the level of free testosterone in the blood.

The British Society for Sexual Medicine has recently released guidelines on male testosterone deficiency,26 and defined normal levels of total testosterone as 12-30nmol/L or a free testosterone level >225pmol/L. As such, it is generally accepted that a total testosterone level >12nmol/L or a free testosterone level >225pmol/L does not require testosterone replacement therapy. In contrast, the guidelines suggest that a total testosterone <8nmol or a free testosterone <180pmol/L will probably require testosterone treatment,24 provided that these results are based on two separate appropriately ordered tests, and other causes have been excluded. A total testosterone level of 8-12nmol/l or a free testosterone level <225pmol/L with symptoms consistent with testosterone deficiency might benefit from a six-month trial of testosterone replacement therapy.

Interestingly, treatment with testosterone therapy aims for a total testosterone level of >15nmol/l, which does pose the question as to whether patients with a total testosterone <15nmol/l and symptoms consistent with TD should be trialled on treatment rather than being subject to the usual cut-off at 12nmol/L.

5. Testosterone replacement therapy does not cause prostate cancer

There are many misconceptions about testosterone replacement therapy (TRT), the most common of which is that it is associated with prostate disease. It is accepted that androgens are essential for healthy prostate function and development, and that use of TRT may stimulate the growth of the prostate. Even so, there is no evidence that patients treated with TRT have an increased risk of developing benign prostatic hypertrophy or that it contributes to worsening of lower urinary tract symptoms.27,28 Furthermore, there is no evidence that TRT increases the risk of developing prostate cancer, providing that malignancy is excluded before starting treatment.4 For patients with suspected prostate or breast cancer, TRT should not be started.

6. Testosterone replacement therapy does not increase cardiovascular risk

Much like the perceived risk of prostate disease with TRT, an increased risk of cardiovascular disease has also been incorrectly associated with TRT. In fact, good evidence now exists that an increased risk of cardiovascular disease is associated with low levels of testosterone, and that normalising levels of testosterone appears to be cardio-protective.5 A large meta-analysis performed by Morgentaler et al. in 20155 found no increase in cardiovascular risk in men receiving TRT and a reduced risk in those patients with metabolic syndrome.

7. Which type of therapy?

Since lifestyle issues play a role in TD (point 3), it is important that any contributing reversible factors, such as obesity, alcohol consumption, or alternative medications, are addressed in conjunction with appropriate prescribing of TRT. However, as with age-related TD, addressing lifestyle factors alone will help – but will not fully correct testosterone levels. Consequently, appropriate TRT becomes an essential component in treating the underlying hormonal deficiency.

In general, TRT can be administered in three ways, with potentially different efficacies and risk profiles. It can be taken orally, applied as a transdermal gel, or given by injection. Other methods include testosterone pellets and subdermal implants (although neither are routine in the UK). All methods are licensed for use in the UK and available through the NHS and private clinics, but there is now a drive to avoid oral therapy where possible due to the high first-pass hepatic metabolism of testosterone.

8. What to expect from testosterone replacement therapy

TRT is a long-term treatment that improves symptoms differentially over time. Patients may experience an increase in sexual interest only after three weeks of treatment, but the effect usually plateaus at about six weeks.6 True changes in erectile function and ejaculation may take up to six months. As to the non-sexual benefits of TRT, improvements in depression and quality of life may be detectable within one month of treatment, but the maximum effect may take up to six months.6 It is for this reason that patients with borderline low testosterone levels and symptoms should be offered a minimum six-month trial of treatment.

9. What are the risks of taking TRT, and how should it be monitored?

The primary aim of TRT is to alleviate the clinical symptoms of TD, and the secondary benefits, such as reduced cardiovascular risk, or increased bone density, are not in themselves an indication to initiate treatment. Due to the risks associated with supra-physiological dosing of testosterone, careful monitoring of changes in the clinical manifestations of TD should be an essential part of every follow-up visit.6 Known risks of TRT include gynaecomastia, acne, erythrocytosis, decreased testicular volume, and peripheral oedema. Perhaps, most importantly, in younger men, TRT also produces infertility. This is because exogenous administration of testosterone acts through a negative feedback mechanism on the hypothalamic-pituitary-gonadal axis to reduce GnRH production, and consequently reduce production of LH and FSH, which are required for natural testosterone and sperm production, respectively. Fortunately, most of these effects reverse when treatment is stopped.

The mechanism of how testosterone links to erythrocytosis is still not entirely clear. It is known that patients on TRT, in particular those in middle age, have an increased risk of developing erythrocytosis, and those on injectable TRT have a higher risk still.29 It has been suggested that TRT stimulates erythropoietin and also increases iron utilisation for erythropoiesis.30 The consequence is a measurable rise in haematocrit. It is therefore essential that during TRT the haematocrit should be kept below 52-55%.7 There is an established link between erythrocytosis and an increased risk of both venous and arterial thrombosis,31 but the risk of erythrocytosis leading to venous thrombosis in TRT is less clear.29 Nevertheless, for patients that do develop erythrocytosis, a haematological assessment is required. Although specialist referral may be indicated, treatment should not be ended abruptly.

TRT should be considered in patients with a total testosterone level of up to 12nmol/L (free testosterone <0.225 nmol/L) and the presence of symptoms.26 The goal of TRT is to return a patient’s serum testosterone level to the upper limit of the normal range, ideally above 15nmol/L,7 and supraphysiological levels should be avoided. As to monitoring patients on TRT, the International Society for Sexual Medicine (in agreement with the British Society for Sexual Medicine) recommends that patients should be reviewed at 3, 6, and 12 months, and then annually.32, 33 Ideally, these reviews should be carried out by the clinician initiating treatment, so as to better gauge the subjective changes that may occur due to TRT, as well documenting objective changes. During reviews, patients should be assessed for the positive effects of TRT, as well as any adverse side effects, medication adherence, and co-morbidity status. In addition, patients should also be examined for changes in weight and central obesity, and any changes in the testes. Contrary to popular opinion, it is low testosterone and not TRT that is associated with the development of hypertension.5 As a consequence, blood pressure should also be recorded during patient reviews, though hypertension relating to testosterone should only be seen in patients where optimal levels of TRT have not been achieved, whether too high or too low. For patients on injectable or gel treatments, administration sites should also be checked.

Although there is a suggestion that a digital rectal examination (DRE) for prostatic changes should be performed annually, it is not evidence-based, and some physicians prefer blood monitoring following the initial exclusion of prostate cancer. Still, the combination of a PSA test and DRE will obviously improve detection rates for prostate cancer. As androgens are essential for normal function of the prostate, it is accepted that TRT will cause a small increase in PSA and prostate volume that stabilises at about 12 months.6 Obviously, a significantly raised PSA or a rising level should warrant a urological referral.

In addition to PSA, blood tests should also include testosterone levels, lipid profile, and haematocrit, as indicated earlier.

10. There is no rule about duration of treatment

After an initial trial of six months’ treatment, patients should only continue with TRT if symptoms have improved. If there is no improvement in libido and sexual performance, improved body fat, reduced fatigue, and muscle function (a relatively subjective measure based on changes in appearance, exercise performance, and daily activities), then treatment should be discontinued. Further investigation of other causes of the patient’s symptoms is then advised. Given the protracted nature of treatment, patients should be counselled that for the vast majority treatment by TRT is long term.


The fundamental principle of testosterone replacement therapy is to return patients’ testosterone from sub-optimal levels to normal physiological levels. Most of the common misconceptions about TRT derive from its association with the effects of supra-physiological doses of the drug and its potential complications, which can be minimised by regular monitoring.

As we live longer, and the numbers of patients with obesity, type 2 diabetes, and other testosterone suppressing conditions, including chronic disease and the use of various medications, increase, it may be time to stop thinking of testosterone deficiency as a normal consequence of ageing in males. Only when we begin to treat TRT as more than a lifestyle drug will we make significant gains in improving the quality of life and mortality for men with this condition.

Dr Jeff Foster, GP Partner, Croft Medical Centre, Warwickshire and TFJ Private GP Services, West Midlands


  1. Hall SA, Esche GR, Araujo AB, Travison TG, Clark RV, Williams RE, McKinlay JB. J Clin Endocrinol Metab 2008 Oct;93(10):3870-7
  2. Bain J. Clin Interv Aging. 2007 Dec; 2(4): 567-576
  3. [Accessed 20/11/2017]
  4. Dohle G, Arver S, Bettocchi C, Kliesch S, Punab M, de Ronde W. European Association of Urology, uroweb: [Accessed November 2017]
  5. Morgentaler A, Miner MM, Caliber M, Guay AT, Khera M, Traish AM. Mayo Clin Proc. 2015 Feb;90(2):224-51
  6. Saad F, Aversa A, Isidori AM, Zafalon L, Zitzmann M, Gooren L. Eur J Endocrinol 2011 Nov;165(5):675-85.
  7. Wylie K, Rees M, Hackett G, Anderson R, Bouloux PM, Cust M, Goldmeier D, Kell P, Terry T, Trinick T, Wu F. [Accessed November 2017]
  8. Hamilton W, Watson J, Round A. BMJ. 2010 Aug 24341:c4259
  9. [Accessed November 2017]
  10. [Accessed November 2017]
  11. [Accessed November 2017]
  12. Kaufman J, Vermeulen A. Endocr Rev 2005 Oct;26(6):833-76.
  13. Wu FC, Tajar A, Pye SR, Silman AJ, Finn JD, O’Neill TW, Bartfai G, Casanueva F, et al. J Clin Endocrinol Metab. 2008 Jul;93(7):2737-45
  14. Mulligan T, Frick MF, Zuraw QC, Stemhagen A, McWhirter C. Int J Clin Pract, 2006 Jul 1; 60(7): 762-769
  15. Felner E, Guillermo E. Umpierrez M. Endocrine Pathophysiology 2013, LWW ISBN 13: 9781451171839
  16. Dandona P, Dhindsa A, Chaudhuri V, Bhatia V, Topiwala S, Mohanty P. Aging Male. 2008. 11(3): 107-117
  17. Exton M, Krüger T, Bursch N, Haake P, Knapp W, Schedlowski M, Hartmann U. World J Urol. 2001 Nov;19(5):377-82
  18. Jannini E, Screponi E, Carosa E, Pepe M, Lo Giudice F, Trimarchi F, Benvenga S. Int J Androl. 1999 Dec;22(6):385-92
  19. Carosa E, Benvenga S, Trimarchi F, Lenzi A, Pepe M, Simonelli C, Jannini E. Int J Impot Res. 2002 Apr;14(2):93-9
  20. Kumagai H, Yoshikawa T, Zempo-Miyaki A, Myoenzono K, Tsujimoto T, Tanaka K, Maeda S. Horm Metab Res. 2017 Sep 21. doi: 10.1055/s-0043-117497
  21. Eliakim A, Nemet D. Harefuah 2006 Sep;145(9):677-81
  22. Safarinejad M, Azma K, Kolahi A. J Endocrinol. 2009 Mar;200(3):259-71
  23. Bhasin S, Cunningham G, Hayes F, Matsumoto A, Snyder P, Swerdloff R, Montori V. Task Force, Endocrine Society. J Clin Endocrinol Metab 2010; 95: 2536-2559
  24. Livingston M, Kalansooriya A, Hartland A, Ramachandran S, Heald A. Int J Clin Pract. 2017 Nov;71(11)
  25. [Accessed November 2017]
  26. Hackett G, Kirby M, Edwards D, Jones T, Wylie K, Osser-Gerning N, David J, Muneer A. Journal of Sexual Medicine, 2017; 14:1504-1523
  27. Bhasin S, Cunningham G, Hayes, F, Matstumoto A, Snyder P, Swerdloff R, Montori V. The Journal of Clinical Endocrinology & Metabolism 2010; 95(6):2536-2559
  28. Dandona P, Rosenberg MT. Int J Clin Pract. 2010 May; 64(6):682-96
  29. Jones S, Dukovac T, Sangkum P, Yafi F, Hellstrom W. Sex Med Rev. 2015 Apr;3(2):101-112
  30. Bachman E, Travison T, Basaria S, Davda M, Guo W, Li M, Connor Westfall J, Bae H, Gordeuk V, Bhasin S. J Gerontol A Biol Sci Med Sci. 2014 Jun;69(6):725-35
  31. Schreijer A, Reitsma P, Cannegieter C. Haematologica. 2010 Feb; 95(2): 182-184
  32. [Accessed November 2017]
  33. [Accessed November 2017].

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read more ...

Privacy & Cookies Policy