Who would benefit from testosterone therapy?
Director of the Centre d’Etude et de Traitement de la Pathologie de l’Appareil Reproducteur et de la Psychosomatique (CETPARP), University of Lille, France
Men with low testosterone levels may present to the healthcare system in a number of different scenarios. Often their symptoms and condition may be overlooked as the association between testosterone deficiency syndrome (TDS) and related co-morbidities (including type 2 diabetes mellitus [DM], metabolic syndrome and cardiovascular disease)1,2 is not appreciated, and the signs and symptoms may be not be specific to TDS. Yet these men may be at an increased risk of mortality compared with men who have normal testosterone levels.3 Screening men at-risk of low testosterone is therefore important, but currently there is no universally agreed threshold for testosterone levels that characterises TDS or at which to commence testosterone replacement therapy (TRT). However, the International Society of Andrology, the International Society for the Study of the Aging Male (ISSAM) and the European Society of Urology have agreed on the following threshold levels for clinical practice:4
- serum total testosterone (TT) levels <8nmol/L (2.31ng/ml) or free testosterone (FT) levels <180pmol/L (52pg/ml) require TRT
- TT levels >12nmol/L (3.46ng/ml) or FT levels >250pmol/L (72pg/ml) do not require TRT
- a trial of TRT may be considered in symptomatic men with TT levels 8-12nmol/L.
These recommendations are helpful in meeting the challenge of diagnosis and management presented by the non-specific nature of TDS. Moreover, if these at-risk patients are tested at their point of entry to the system and managed systematically it can improve not only their TDS, but may also improve their associated co-morbidities.1,2
The association between TDS and other disorders
Estimates of TDS prevalence differ according to the thresholds used, and the criteria for presence or absence of symptoms. Perhaps the best estimate in men over 40 years was made by Araujo et al from the Massachusetts Male Aging Study (MMAS), based on the following operational definition: at least 3 of 8 signs and symptoms of hypogonadism, plus TT <6.94nmol/l (2ng/ml), or TT <13.88nmol/l (4ng/ml) and FT < 0.31nmol (89pg/ml).5 Crude prevalence of TDS was 6%.5
TDS is often associated with other conditions, most notably type 2 DM, metabolic syndrome, cardiovascular disease and erectile dysfunction (ED). Cross-sectional studies have repeatedly demonstrated that men with type 2 DM are significantly more likely to have low testosterone levels. The estimated prevalence from numerous studies varies from 20% to 64% depending on the population and which testosterone measure was used (TT, FT or bioavailable).6,7–10 In a large non-institutionalised population of men aged over 20 years, men in their lowest tertile of FT or BT were four times as likely to have prevalent diabetes.11
Men with metabolic syndrome, defined by insulin resistance, obesity, lipid abnormalities and hypertension, are also at increased risk of TDS. In a study involving 803 men with sexual dysfunction (mean age of 53.6 years), circulating TT was <8nmol/L (2.31ng/ml) in 11.9% of those with metabolic syndrome compared with 3.8% in the rest of the population, and this decreased as the number of metabolic syndrome components present increased.12 In a separate cross-sectional study of 400 men aged 40–80 years, TT and sex hormone binding globulin (SHBG) were inversely related to insulin resistance and to metabolic syndrome.13
There is evidence that low testosterone levels, is by itself a risk factor for diabetes and the metabolic syndrome. Both the Rancho Bernardo and MMAS studies found a significant inverse relationship between a low level of FT and TT, with a risk of developing diabetes after 8 and 9 years, respectively (OR: 2.7 in the Rancho-Bernardo study).14,15
An assessment of cross-sectional studies has shown a significantly higher prevalence of coronary artery disease (CAD) in the presence of low testosterone levels in 16 out of 32 trials reviewed. However, none of the seven prospective cohort studies or nested control studies showed any association between lower (or elevated) levels of testosterone and symptoms of CAD.16 In MMAS, the risk for having a TT <10.4nmol/L was significantly higher in men with vascular risk factors (odds ratios: obesity: 2.38, diabetes: 2.09, hyperlipidaemia: 1.47, hypertension: 1.84), than in men without these conditions.5 Lastly in a recent retrospective study of 858 male veterans followed for 8 years, men with a repeatedly low TT (<8.7nmol/L [2.5ng/ml]) or FT (<0.03nmol/L [75pg/ml]) level had a significantly increased mortality even after exclusion of prostate cancer, of subjects with first year mortality, and after multivariate adjustment (OR: 1.65).3
The prevalence of low testosterone in men with ED has been estimated at 10-20%.17 A review of nine large studies with routine determination of TT in patients consulting for ED found testosterone levels <10.4nmol/L (3ng/ml) in 14.7% of the 4342 men over 50 years of age.18
Low testosterone levels are also known to be linked with conditions such as osteoporosis. An assessment of TT in men over 65 years of age found a prevalence of marked testosterone deficiency (<6.9nmol/L) twice as high in osteoporotic men as in those with normal bone mineral density (BMD). The incidence of rapid hip-bone loss was also significantly higher in men with a marked TT deficiency than in those with normal TT (22.5% and 8.6%, respectively; p=0.007).19
Signs and symptoms of TDS
The clinical picture of TDS, defined as both low testosterone levels and clinically significant symptoms, is non-specific and therefore there is no definitive set of symptoms or profile.20,21 Symptoms of TDS not only present with TT levels below 8nmol/L, but may also occur at various TT levels between 8-12nmol/L, depending on the type of TDS symptom and on the individual sensitivity to testosterone of each man.7,22 Established TDS may include several of the following:, reduced muscle mass and strength, reduced energy, poor concentration, reduced sexual desire and ED, and metabolic disturbances.20,21,23
In patients receiving TRT in the form of subcutaneous implants, the first symptoms associated with recurrence of TDS due to expiry of the implant are lack of energy and of libido.22 Another indication that these two symptoms are the most sensitive markers for TDS is shown by a study of ageing males – among seven TDS symptoms associated with a significantly lower testosterone concentration, lack of libido and lack of vigour were those whose prevalence significantly increased below the higher testosterone concentrations (<15nmol/L [4.33ng/ml]), while the prevalence of disturbed sleep, feeling depressed and lacking concentration significantly increased <10nmol/L (2.88ng/ml) and <8nmol/L (2.31ng/ml) for hot flushes and ED. Upon physical examination, reduced body and facial hair, small and soft testes, and gynaecomastia are indicative of low testosterone levels, but are not present in every TDS patient.7 Table 1 provides an overview of those symptoms most indicative of TDS and others commonly associated, but which are less specific.
Table 1. Signs and symptoms indicative of TDS4,23
|Most specific signs and symptoms||Less specific signs and symptoms||Conditions associated with a high prevalence of low testosterone levels|
Clinical benefits associated with TRT
A literature review identified over 60 randomised clinical trials of TRT in ageing men. These RCTs persuasively demonstrate beneficial effects of TRT in men over 40 years of age. These include improvements in body composition (increase in lean mass and decrease in fat mass), bone mineral density, sexual desire and CAD. Positive results have also been observed in some studies in muscle strength, cognition and well-being, erectile function and sexual activity [Table 2].1,2,18,24,26–31 It must be emphasised that the majority of these trials were of short duration (less than a fifth lasted more than 6 months) and many studies were insufficiently powered to demonstrate absence of efficacy. The benefit of treatment was ascribed to TRT only if it was statistically significantly greater than the placebo effect.
Table 2: Effects of TRT
|Effect of TRT|
|Cognition, mood, energy and well being26||Four of the nine trials which assessed cognitive abilities found statistically significant beneficial effects on:|
However, no significant effect over placebo was found on overall memory, recall, or verbal fluency.26
TRT also significantly increased energy or decreased fatigue (3/5 trials), and improved some measures of well being or quality of life (6/12 trials). Conversely, only three of the 15 studies which assessed the effects of TRT on mood in ageing men found a statistically significant improvement in depression or other mood scores compared with placebo.26
|Body composition27||The most consistently beneficial effects of TRT were changes in body composition:27|
|Diabetes1,2||In an open-label, randomised, no-treatment controlled study of 48 patients with type 2 diabetes over 45 years of age, TRT significantly decreased waist-to-hip ratio, fasting and post-prandial glucose, HbA1c (from 10.4% to 8.6%) as well as clinical symptoms of TDS, which did not change in the control group.1|
In addition, in a double-blind, placebo-controlled crossover study (two 12 week periods) of TRT in 24 men over 50 years of age with type 2 diabetes mellitus and TT <12nmol/L (3.46ng/ml), TRT resulted in significant improvements in waist circumference, waist-to-hip ratio, blood fasting glucose, insulin sensitivity (HOMA), HbA1c and cholesterol.2
|Coronary artery disease24,28||Two studies have shown a direct vasodilatory action of pharmacological concentrations of testosterone on coronary arteries in human males (in vivo),24 and short-term improvements in ECG changes of CAD during exercise-tests (increased time to ischemia) in four placebo-controlled trials of TRT.24,28|
|Bone mineral density1,26,27||Six trials of 12 to 36 months’ duration studied the effects of TRT on men over 50 years of age. The group receiving TRT responded significantly better than the placebo group (5/6 trials).26,27 In a meta-analysis of these trials, pooled effects revealed that treatment effect was higher at the lumbar spine and femoral neck sites.1 However, due to high baseline variability of the study population, only the estimated increase in bone mineral density at the lumbar spine site (3.7%) reached statistical significance.1|
Until now no adequately powered interventional trials have explored the outcome of TRT administration on rates of long bone and vertebral fractures in men of any age.
|Sexual function18,27,29,30||A meta-analysis integrated 17 trials investigating TRT on sexual function of men of any age (mean age 57.5 years, >50 years of age in 10 of the 17 RCTs). A large and significant effect on all aspects of sexual function, including sexual motivation, sexual thoughts, erectile function, and the frequency of successful intercourses was found in men with low (<7nmol/L [2ng/ml]) and low–normal TT (7nmol/L–12nmol/L [2ng/ml–3.46ng/ml]) with respect to placebo. No statistically significant effect was found in trials with mean baseline TT levels above 12nmol/L (3.46ng/ml).27 The effect of age was not studied.|
When specifically looking at the 18 RCTs that assessed the effects of TT on sexual function of men over 50 years of age, libido significantly increased in eight of the 12 studies.26 TRT also increased the frequency of sexual activity in three out of five studies, and improved erectile function over placebo in a further seven of the 16 studies. The latter included two trials of ED patients with low or low–normal testosterone who were non-responders to sildenafil (phosphodiesterase-5 inhibitor [PDE5i]).29,30 The combination of this with TRT significantly improved its effectiveness with respect to its combination with placebo. These data confirm the beneficial effect of TRT as treatment of sexual dysfunction in men over 40 years of age. The effect on libido predominates, while that on erectile function is less significant in men aged over 40 years. TRT may thus be used as a monotherapy, or as part of a combined therapy with a PDE5i when it is not able to restore erections alone due to the frequent association of vascular factors with testosterone deficiency in older men with ED.18
|TDS in men with systemic diseases31||TRT has a beneficial effect in:31|
Open studies have also shown positive effects (but these have yet to be confirmed in placebo controlled studies) in:
Assessment of patients with suspected TDS: practical recommendations for patient selection and screening
Who should I select for biochemical assessment?
Unfortunately no one sign or symptom is absolutely specific for TDS, and many are very common. Different scales grouping together these signs and symptoms have been proposed to aid diagnosis (ADAM questionnaire, Aging Male Symptoms [AMS] scale) 21,23 and in clinical practice they have a role in encouraging patients to discuss their symptoms and in monitoring changes in symptoms.
Men with ED, reduced libido, decrease in muscle mass and strength, type 2 DM and metabolic syndrome should especially be assessed for TDS.32 Measurement of testosterone is also recommended under the following circumstances where there is a high prevalence of low testosterone levels: treatment with medications that alter testosterone production or metabolism (e.g. glucocorticoids, opioids, ketoconazole, chemotherapy), HIV-associated cachexia, end-stage renal disease and chronic obstructive pulmonary disease.23 However, measurement should be avoided at the time of acute disease or of acute aggravation of a chronic disease, as testosterone secretion may be transiently inhibited.
Biochemical assays and other tests to be conducted
The biochemical diagnosis of TDS is based on the measurement of serum TT, preferably before 11:00am, though the diurnal rhythm of testosterone is less marked in men over 40 years of age. Serum FT or BT are theoretically more representative of the biological activity of testosterone, and can also be assayed, but most authors do not recommend them as first measurement. Indeed these assays may be difficult to carry out (especially equilibrium dialysis, the reference method for FT), inaccurate (FT assay by testosterone analogue and ‘Free Androgen Index’), or available in few laboratories (BT). In addition reasons to prefer them to the simpler TT assay have yet to be established in clinical practice.21 These assays are generally set aside for the repeat assay, and then used to clarify the significance of a borderline level of TT, or when SHBG concentration may be altered. Presently, most physicians are favouring the ‘calculated FT’ (cFT), which is calculated from TT and SHBG according to a specific formula (a calculator is available on the ISSAM website at www.issam.ch).5,21
In case of low or borderline testosterone value, the assay should be repeated, because of the frequent intra-individual fluctuations of serum testosterone, unless physical evidence of hypogonadism (i.e. atrophied testes) is present. The assays of serum luteinising hormone (LH) and follicle stimulating hormone (FSH) (to specify the primary/testicular or secondary/hypothalamic or pituitary origin of hypogonadism), prolactin (to screen for hyperprolactinaemia, a major cause of low testosterone), and possibly SHBG (to calculate cFT) should be combined with the repeat testosterone assay.
Confirmation of a secondary hypogonadism (by low or normal serum LH and FSH) should lead to the measurement of serum prolactin, iron saturation (screening for haemochromatosis), and if hypogonadism is marked (testosterone <7nmol/L [2ng/ml]), or associated with persistent hyperprolactinaemia, panhypopituitarism, or symptoms of tumour mass such as headache, visual impairment or visual field defect, to pituitary imaging with magnetic resonance imaging. Such diagnoses would imply specific precautions or treatments
Initiating testosterone replacement therapy in patients with TDS
Contraindications and warnings
TRT is indicated when the combination of a consistently low testosterone level on at least two occasions, and clinically significant symptoms compatible with testosterone deficiency are present.4, 20,21,23 Contraindications to TRT include a history of breast or prostate cancer, a palpable prostate nodule or prostate serum antigen (PSA) >10.4nmol/L without urological evaluation, severe low urinary tract symptoms, a polycythaemia and uncontrolled severe congestive heart failure. Men with obstructive sleep apnoea syndrome and a prolactinoma with uncontrolled hyperprolactinaemia must receive TRT with caution. As the possible development of a contraindication during treatment (especially prostate carcinoma) requires rapid discontinuation of TRT, it is recommended that short-acting (transdermal, oral or buccal) preparations rather than long-acting (intramuscular, subdermal) depot preparations be used in older men.4
Testosterone therapy test
There is a consensus that testosterone levels <8nmol/L (2.31ng/ml) are an indication for initiation of TRT. In such patients, long-term treatment can be scheduled from the outset.4 For symptomatic men with borderline or low/normal testosterone levels (8–12nmol/L [2.31–3.46ng/ml]) a ‘testosterone therapy test’ may be considered. There is not always a clear causal relationship between the low/normal testosterone levels and associated symptoms. However, several meta-analyses of RCTs have established that significant improvements may be achieved with TRT in men with circulating testosterone levels in this range, especially with respect to body composition and sexual function.1,29,33 This justifies the ‘testosterone therapy test’ 4,20 for a limited period. Following this, TRT is continued only if the expected improvements have been obtained. Since clinical benefit may not become apparent for 3 months or longer, for ED34,35 the ‘testosterone therapy test’ should be carried out for at least 3 months. Patients should be seen again after 2–3 months of treatment, and encouraged to continue to the end of the trial. In cases of TRT for ED with low testosterone levels, the possibility to extend the test up to 6 months could be offered in case of insufficient effectiveness at 3 months.
To be interpretable the ‘testosterone therapy test’ must actually raise the circulating testosterone level into the mid-normal range for men under 50 years of age. The obtained level must therefore be measured 2–4 weeks after initiation of the test, and the testosterone product must be titrated if the testosterone concentration is not in the expected range. On the other hand, symptoms must be assessed as objectively as possible, using validated scales such as the AMS (which has a good sensitivity to assess the effects of TRT)36 and the International Index of Erectile Function for the assessment of sexual symptoms.
Both the physiological effects of treatment and the testosterone levels achieved should be monitored.4, 20,21,23 The clinical parameters most sensitive to abnormally low testosterone levels are physical and mental vigour/energy, well being, libido and sexual thoughts, and nocturnal and morning erections. The first effects of TRT may be perceived within 2–4 weeks, but, for sexual function, effects may sometimes take 3–6 months to become apparent34 and even up to 12 months for the nocturnal erections reaching the normal range in previously untreated hypogonadal patients.35 In the case of osteoporosis, bone mineral density should be checked only after 1-2 years of therapy.
Assessment of safety is mainly based on clinical assessment (especially weight, blood pressure, digital rectal examinations) and repeat PSA and haematocrit measurements.4,20,21,23 Urological consultation becomes mandatory in case of a verified PSA concentration >4ng/ml, increase in PSA concentration >1.4ng/ml within any 12-month period of treatment, PSA velocity of >0.4ng/ml per year, using the PSA level after 6 months of TRT as the reference, of detection of a prostate abnormality on digital rectal examination, and of an American Urological Association or International prostate symptom score >19.23 In the case of a haematocrit >54%, TRT should be stopped until normalisation, then reinitiated at a lower dose.
Monitoring usually starts 3 months after initiation of therapy, with clinical, testosterone levels, haematocrit, and after age 40 PSA assessments. Then prostate assessments are recommended quarterly for 1 year, then at 6-month intervals as with the clinical, testosterone and haematocrit assessments.
It is important for physicians to consider testing the testosterone levels of at-risk men aged over 40 years seen in all clinical settings for TDS, and especially those with type 2 DM, metabolic syndrome, cardiovascular disease and ED, as these groups are at high risk, but are currently under-recognised and under-diagnosed. The combination of typical clinical signs and symptoms, together with treatment history can help to identify candidates for testosterone screening and potentially TRT, which may include a testosterone therapy test. As seen, treatment with TRT can lead to significant improvements in patients’ quality of life, and may also improve outcomes in certain associated disorders.
- Boyanov MA, Boneva Z, Christov VG. Testosterone supplementation in men with type 2 diabetes, visceral obesity and partial androgen deficiency. Aging Male 2003; 6(1): 1–7.
- Kapoor D, Goodwin E, Channer KS, Jones TH. Testosterone replacement therapy improves insulin resistance, glycaemic control, visceral adiposity and hypercholesterolaemia in hypogonadal men with type 2 diabetes. Euro J Endocrinol 2006; 154: 899–906.
- Shores MM, Matsumoto AM, Sloan KL, Kivlahan DR. Low serum testosterone and mortality in male veterans. Arch Intern Med 2006; 166: 1660–1665.
- Nieschlag E, Swerdloff R, Behre HM et al. Investigation, treatment and monitoring of late-onset hypogonadism in males. ISA, ISSAM, and EAU recommendations. Eur Urol 2005; 48: 1–4.
- Araujo AB, O’Donnell AB, Brambilla DJ et al. Prevalence and incidence of androgen deficiency in middle-aged and older men: estimates from the Massachusetts Male Aging Study. J Clin Endocrinol Metab 2004; 89: 5920–5926.
- Mulligan T, Frick MF, Zuraw QC et al. Prevalence of hypogonadism in males aged at least 45 years: the HIM study. Int J Clin Pract 2006; 60: 762–769.
- Zitzmann M, Faber S, Nieschlag E. Association of specific symptoms and metabolic risks with serum testosterone in older men. J Clin Endocrinol Metab 2006; 91: 4335–4343.
- Kalyani RR, Dobs AS. Androgen deficiency, diabetes and the metabolic syndrome. Curr Opin Endocrinol Diabetes Obes 2007; 14: 226–234.
- Dhindsa S, Prabhaka S, Sethi M et al. Frequent occurrence of hypogonadotropic hypogonadism in type 2 diabetes. J Clin Endocrinol Metab 2004; 89: 5462–5468.
- Corona G , Mannucci E, Petrone L et al. Association of hypogonadism and type II diabetes in men attending an outpatient erectile dysfunction clinic. Int J Impot Res 2006; 18: 190–197.
- Selvin E, Feinleib M, Zhang L et al. Androgens and diabetes in men: results from the Third National Health and Nutrition Examination Survey (NHANES III). Diabetes Care 2007; 30: 234–238.
- Corona G, Manucci E, Schulman C et al. Psychobiological correlates of the metabolic syndrome and associated sexual dysfunction. Eur Urol 2006; 50: 595–604.
- Muller M, Grobbee DE, den Tonkelaas I et al. Endogenous sex hormones and metabolic syndrome in aging men. J Clin Endocrinol Metab 2005; 90: 2618–2623.
- Oh J, Barrett-Connor E, Wedic NM, Wingard DL. Endogenous sex hormones and the development of type 2 diabetes in older men and women, the Rancho Bernardo Study. Diabetes Care 2002; 25: 55–60.
- Stellato RK, Feldman HA, Hamdy O et al. Testosterone, sex hormone-binding globulin, and the development of type 2 diabetes in middle-aged men: prospective results from the Massachusetts male aging study. Diabetes Care 2000; 23: 490–494.
- Wu FC, von Eckardstein A. Androgens and coronary artery disease. Endocr Rev 2003; 24: 183–217.
- Roumequère T. Rationale for androgens and erectile dysfunction in 2006. Eur Urol 2006; 50: 898-900.
- Buvat J, Boujaoude G. Significance of hypogonadism in erectile dysfunction. World J Urol 2006; 24: 657–667.
- Fink HA, Ewing SK, Barrett-Connor E et al. Association of testosterone and estradiol deficiency with osteoporosis and rapid bone loss in older men. J Clin Endocrinol Metab 2006; 91: 3908–3915.
- Nieschlag E, Behre HM. Clinical uses of testosterone in hypogonadism and other conditions. In Testosterone, Action, Deficiency, Substitution (Nieschlag E and Behre HM eds). Cambridge University Press, 3rd edition, 2004, pp 375–404.
- Buvat J, Shabsigh R, Guay A et al. Hormones, Metabolism, Aging and Men’s Health. In Standard Practice in Sexual Medicine (Porst H and Buvat J eds). Blackwell Publishing, 2006, pp 225–286
- Kelleher S, Conway AJ, Handelsman DJ. Blood testosterone threshold for androgen deficiency symptoms. J Clin Endocrinol Metab 2004; 89: 3813–3817.
- Bhasin S, Cunningham GR, Hayes FJ et al. Testosterone therapy in adult men with androgen deficiency syndromes: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 2006; 91: 1995–2010.
- Gruenewald DA, Matsumoto AM. Testosterone supplementation therapy for older men: potential benefits and risks. J Am Geriatr Soc 2003; 51: 101–115.
- Testosterone and Aging. Clinical Research Directions. Committee on assessing the need for clinical trials of testosterone replacement therapy. Institute of Medicine of the national academies. (CT Liverman, DG Glazer eds). The National Academic Press, Washington DC.
- Buvat J, Boujaoudé G. Testosterone replacement therapy in the aging man. J Men’s Health Gender 2005; 2: 396–399.
- Isidori AM, Gianetta E, Greco EA et al. Effects of testosterone on body composition, bone metabolism and serum lipid profile in middle-aged men: a meta-analysis. Clin Endocrinol 2005; 63: 280–293.
- Liu PY, Death AK, Handelsman DJ. Androgens and cardiovascular disease.
Endocr Rev 2003; 24: 313–340.
- Aversa A, Isidori AM, Spera G et al. Androgens improve cavernous vasodilation and response to sildenafil in patients with erectile dysfunction. Clin Endocrinol (Oxf) 2003; 58: 632–638.
- Shabsigh R, Kaufman JM et al. Randomized study of testosterone gel as adjunctive therapy to sildenafil in hypogonadal men with erectile dysfunction who do not respond to sildenafil alone. J Urol 2004; 172: 658–663.
- Kalyani RR, Gavini S, Dobs AS. Male hypogonadism in systemic disease. Endocrinol Metab Clin North Am 2007; 36: 333-348
- Channer K, Dumas C. Prospective evaluation of the effects of testosterone therapy in hypogonadal men with type 2 diabetes or metabolic syndrome: demographic analysis of TIMES2 study. Diabetes Vasc Dis Res 2007: 4: S164.
- Boloña ER, Uraga MV, Haddad RM et al. Testosterone use in men with sexual dysfunction: a systematic review and meta-analysis of randomized placebo-controlled trials. Mayo Clin Proc 2007; 82: 20–28.
- Yassin A, Saad F. Improvement of sexual functions in men with late onset hypogonadism treated with testosterone only. J Sex Med 2007; 4: 497–501.