What is known
Diagnosis and treatment of testosterone deficiency is a complex issue, for many reasons. Symptoms that may be indicative of testosterone deficiency are not specific to this condition, and may be related to co-morbidities.2,3 Despite proposed testosterone thresholds by clinical guidelines (which are not universally agreed upon) there is no single testosterone level – which applies to all men - at which symptoms and signs indicative of testosterone deficiency start to develop.4,5
Further complicating the issue of diagnostic testosterone thresholds is the tremendous variability in testosterone assays – each giving different readings - and assay specific reference ranges that vary between labs.6,7 In addition, because of differences in androgen receptor sensitivity, among men with the same testosterone level, some may suffer hypogonadal symptoms, while others don’t.8
Adding to this complexity are the multiple available testosterone treatment options.9,10 Available testosterone preparations include transdermal, intramuscular injections, oral, sublingual/buccal and pellet implants. Each has its own pros and cons that need to be considered when treating hypogonadal men.
What this study adds
The study by Carruthers et al. reports clinical experience with seven different testosterone preparations available in the UK.11 The goal was to investigate symptom response to testosterone treatment with testosterone undecanoate injections, pellet implants, scrotally applied gel, scrotally applied cream, transdermal testosterone gel, oral testosterone undecanoate or mesterolone. In addition, effects on endocrine, biochemical and physiological responses to the different preparations were investigated.
2247 men, mean age 54 years (range 20-90 years) attending the 3 Men’s Health Centers were treated with one of seven different testosterone preparations.
A notable aspect of this study is that diagnosis was made on the basis of clinical signs and symptoms alone – regardless of baseline testosterone levels - and it included many patients who had previously been declined treatment by their doctors because they were deemed to have testosterone levels in the "normal" range. Overall, only 2% of pre-treatment total testosterone levels were below 6 nmol/l (173 ng/dl), 6% below 8 nmol/l (230 ng/dl), 17% below 10 nmol/l (288 ng/dl) and 31% below 12 nmol/l (346 ng/dl). These are all commonly used thresholds required for the diagnosis of hypogonadism, as proposed by several clinical guidelines.3,12,13
Correlation of baseline testosterone levels with symptom severity and symptom resolution
Surprisingly to many, it was found that baseline total testosterone and free testosterone (calculated) levels were totally unrelated to the initial symptom score (which was the reason for patients attending the clinic for treatment). Also, there was no association between initial symptom score and baseline estradiol or SHBG. This is consistent with results from other studies showing failure of hormone parameters to correlate with testosterone deficiency symptoms.14,15 This clearly shows that sole reliance on testosterone levels is inappropriate when evaluating men for testosterone deficiency.
Comparing the symptomatic responses to treatment in the patients with initial total testosterone levels below 12 nmol/l (mean 9.6 nmol/l) with those greater than or equal to 12 nmol/l (mean 19.1 nmol/l), there was no significant difference between either the initial symptom scores or their response to treatment for up to 12 years, regardless of testosterone preparation used for treatment. This can be explained by inter-individual differences in androgen receptor sensitivity, which makes some men resistant to the effects of testosterone.8 Those men can have markedly higher testosterone levels but still experience the same degree of symptoms as do men with much lower levels.
Symptom resolution as a function of testosterone preparation and treatment duration
Comparing the different testosterone preparations, the rates for failure to achieve complete symptom response at 1 year with testosterone undecanoate injections, pellet implants, scrotally applied gel, testosterone gel, scrotal cream, oral testosterone undecanoate and mesterolone were 18, 23, 21, 35, 35, 49 and 59%. Thus, within the first year of treatment, testosterone undecanoate injections performed very well in terms of symptom resolution.
However, among patients who did not achieve resolution of symptoms during the first year of testosterone treatment, those who stayed on treatment eventually did achieve significant symptomatic benefit compared to baseline. Thus, even subjective symptoms (not only biochemical parameters and cardiometabolic risk factors) may take longer than 1 year to resolve.
This is an important finding, as symptomatic relief is often noticeable within 3 to 6 months after start of testosterone therapy16 and many physicians discontinue testosterone treatment in patients who do not experience symptomatic improvement within this time period. It also suggests that the previously suggested duration for a testosterone treatment trial of 3 months in symptomatic men17 may not be enough to allow for symptomatic relief.
Safety of the different testosterone preparations
The safety factors monitored using these preparations, including renal and liver function tests, remained normal. Also, there was no clinical evidence of either increased cardiovascular events or venous thrombosis. Cardiovascular risk factors either did not change or improved (reduction in cholesterol and diastolic blood pressure).
In few cases, treatment with pellet implants or testosterone gel caused excessive increases in hemoglobin or hematocrit and polycythemia. However, this was reversed by phlebotomy, dose reduction or switching to another preparation. Regardless of testosterone preparation, there were no excessive increases in PSA nor any adverse prostate changes, and the incidence of prostate cancer was equivalent to that in the general population, as had been reported previously.18
Key findings and conclusion
This study highlights two important issues relating to diagnosis and treatment of testosterone deficiency:
- Symptoms indicative of testosterone deficiency do not correlate with either total or free testosterone levels a baseline. This contradicts clinical practice guidelines which require a combination of symptoms and "unequivocally low testosterone levels" to make the diagnosis of testosterone deficiency3,12,13, but supports the view of other researchers that initiation of testosterone treatment should rely more on symptoms and less on a discrete numerical testosterone level thresholds.19
- Symptomatic relief may require longer than 1 year to achieve. This may contribute to the low compliance seen with even the use of the very popular transdermal testosterone gels: only 31% and 14% of patients remain on therapy for 6 months and 1 year, respectively.20 Therefore, the importance of testosterone treatment adherence and patient education should be underscored in clinical practice.
In particular, this study gives further evidence against denying men with obvious symptoms testosterone therapy testosterone therapy because they have "normal" testosterone levels.11 Due to excessive reliance on laboratory measures of testosterone levels and inappropriate safety concerns, many men who could greatly benefit from symptom relief and improvement in risk factors remain untreated and deprived of clinically significant preventive medical benefits. This not only causes unnecessary personal suffering, but also imposes a marked financial cost upon an already burdened health care system.21,22