Ju Tae Seo, Department of Urology, Cheil Hospital, Sungkyunkwan University School of Medicine, Seoul, Korea. e-mail: firstname.lastname@example.org
To measure bone mineral density (BMD) and testosterone levels in patients with non-mosaic Klinefelter’s syndrome (KS), as a low BMD is common in hypogonadal men, but little is known about the relationship between BMD and serum testosterone levels in men with KS.
PATIENTS, SUBJECTS AND METHODS
The study included 40 patients with KS and 20 healthy fertile men recruited as controls. Serum testosterone, follicle-stimulating hormone (FSH) and luteinizing hormone (LH) levels were measured by radioimmunoassay. The BMD was obtained at the lumbar spine (L2–4), femoral neck and Ward’s triangle, by dual-energy X-ray absorptiometry.
The serum testosterone level was lower, while FSH and LH were higher, in patients with KS than in the control group (P < 0.001). Patients with KS had a significantly lower mean (sd) BMD at the lumbar spine, femoral neck and Ward’s triangle, than the controls, at 0.88 (0.13) vs 1.09 (0.10) (P < 0.001), 0.78 (0.12) vs 0.87 (0.10) (P = 0.006) and 0.65 (0.12) vs 0.76 (0.11) g/cm2 (P = 0.001), respectively. There was a linear correlation between all BMD values and serum testosterone levels in men with KS who had a low serum testosterone level.
There was a relationship between a lower BMD and testosterone levels in patients with KS. These findings suggest that low testosterone levels cause inadequate bone development and a low BMD in men with KS; therefore, testosterone replacement might be necessary to prevent bone mineral deficiency and future risk of fractures in men with KS who also have low serum testosterone levels.
The WHO definition of osteoporosis for postmenopausal Caucasian women, i.e. a bone mineral density (BMD) of <2.5 sd below the mean for young adult women (age 30 years), was developed for use in studies of that specific population . However, this standard has been applied to individuals of different races and to men, with no thorough investigation of the utility and validity of these standards for these other groups. According to the WHO definition of osteoporosis, osteoporosis is less prevalent in men than in women, possibly due in part to a man’s greater body mass, greater bone size, greater accrual of bone during growth, absence of a clear decrease in endogenous sex hormones analogous to the menopause, and shorter average lifespan than women .
Osteoporosis is reported to affect increasingly many men worldwide, causes significant morbidity and mortality, and significant health and social service expenditure [3,4]. About 15% of all vertebral compression fractures and 20–25% of hip fractures occur in men [5,6]. In a European study, the incidence of osteoporosis at the femoral neck, in subjects aged >50 years, was 6% in men and 23% in women . At present it is thought that ≈ 30% of hip fractures occur in men ; the mortality during the month after such fractures was reported to be higher in men than in women, and estimated to be 10–14%. A major cause for male osteoporosis was identified in only 30–50% of cases, depending on the series reported [9,10].
Although idiopathic and involutional osteoporosis are common diagnoses, secondary causes of a diminished BMD, including hypogonadism, glucocorticoid excess, alcoholism, hypercalciuria, malabsorption and hyperthyroidism, are also frequent in men referred with a diagnosis of osteoporosis .
Klinefelter’s syndrome (KS) is a common sex chromosome disorder that involves an extra X chromosome in males; it is found in 1 in 500 males, and results in impaired spermatogenesis and androgen deficiency. When Klinefelter et al. originally described this syndrome there was no reference to osteoporosis. Several studies since the initial report of KS found that hypogonadism in patients with KS is associated with an accelerated rate of bone loss and osteoporosis [13,14]. We noted that the BMD is often decreased in hypogonadal males and therefore we assessed whether the BMD correlates with testosterone levels in patients with KS.
PATIENTS, SUBJECTS AND METHODS
From January 2001 to May 2004, 40 patients with chromatin-positive KS (mean age 32.05 years, sd 3.10) had a careful medical history taken and the accuracy of data verified by checking all available medical records. The diagnosis of KS was based on a chromosomal study; the karyotype of all patients was 47XXY, and men with chromosomal mosaicism were not included. None of the patients used medication known to affect the skeleton (e.g. testosterone, growth hormone, calcium or vitamin D). Venous blood samples were collected between 08:00 and 10:00 hours, and were stored at − 20 °C until assayed.
Twenty age-matched men were recruited as controls; no patient or control subject had a significant medical illness or was taking calcium supplementation or any other drug that could affect bone metabolism. The control group was recruited mainly among physicians in the authors’ hospital, and less often among patients who attended the clinic for disorders unrelated to bone metabolism. None of the controls had risk factors for osteoporosis; in particular, they did not smoke, their alcohol intake was less than one glass of wine per day, they had no history of low-energy fractures, and they had no osteoporosis as measured by bone densitometry, whether in the spine or at the femoral neck or Ward’s triangle (T score >− 2; the T score is the number of sds from the mean BMD of an average 30-year-old). Institutional Review Board approval and informed consent were obtained.
Bone densitometry (BMD; the area density in g/cm2) was assessed using dual-energy X-ray absorptiometry operating in the fan-beam mode (QDR-2000 densitometer, Hologic, Inc., Waltham, MA, USA). Quality control scans were taken daily, using an anthropometric spine phantom supplied by the manufacturer; the long-term (>1 year) coefficient of variation was 0.40%. Lumbar BMD was assessed at L2–4 on the basis of the postero-anterior view. Fractured vertebrae were excluded from analysis. The femur neck and Ward’s triangle BMD was measured at the upper left femur (Fig. 1). The mean precision error of absorptiometry measurements was <1.5% for the lumbar spine, and <2% for the femur neck and Ward’s triangle BMD. T scores were calculated using the manufacturer’s references.
The results are expressed as the mean (sd); the results for patients with KS and controls were compared using Student’s t-test, with anova if needed. Linear regression analysis was used to determine coefficients of correlation; the test used was Pearson’s r, with P < 0.05 considered to indicate statistical significance.
The mean serum testosterone level was 1.89 (1.40) nmol/L, the mean serum LH was 13.74 (4.40) IU/L and the mean serum FSH was 36.92 (10.26) IU/L. Serum testosterone levels were lower, and FSH and LH levels higher, in patients with KS than in the control group (P < 0.001). The mean level of serum testosterone for those with KS was 1.63 (1.09) ng/mL and for controls it was 5.27 (1.14) ng/mL (Table 1).
Table 1. Anthropometric characteristics, hormonal assays and a comparison of regional BMD in patients with KS and in controls
Mean (sd, range)
KS (40 men)
Controls (20 men)
32.05 (3.10, 27–37)
31.75 (2.77, 26–36)
176.93 (4.70, 167–186)
172.90 (5.07, 165–183)
76.85 (11.93, 57–110)
73.50 (7.04, 60–88)
1.89 (1.40, 0.1–4.1)
5.27 (1.14, 4.1–7.5)
4.17 (1.64, 2.0–8.4)
36.92 (10.26, 20–61)
4.89 (1.99, 2.1–11.3)
0.88 (0.13, 0.67–1.17)
1.09 (0.10, 0.93–1.24)
L2–4, Z score
−2.23 (1.20, − 4.03 to 0.51)
−0.26 (0.89, − 1.73 to 1.16)
0.78 (0.12, 0.59–1.11)
0.87 (0.10, 0.68–1.07)
Neck Z score
−1.24 (1.05, − 3.14 to 1.72)
−0.47 (1.01, − 2.27 to 1.43)
Ward’s triangle, g/cm2
0.65 (0.12, 0.43–0.90)
0.76 (0.11, 0.57–0.98)
Ward Z score
−0.95 (1.05, − 3.58 to 1.06)
0.05 (0.96, − 1.56 to 1.86)
Patients with KS had a significantly lower BMD at the lumbar spine, femoral neck and Ward’s triangle than the controls (Table 1); the mean Z score (the number of sds from the mean of an age-, gender- and ethnically-matched reference group) was − 2.23 for the lumbar spine, − 1.24 for the femoral neck and − 0.95 for Ward’s triangle. All BMD values were almost ≥ 1 sd below the reference mean in patients with KS (Table 1).
There was a statistically significant linear correlation between the BMD or Z score of the spine and baseline testosterone levels (Fig. 2A,B) in all patients with KS. There was a weak but statistically significant inverse correlation between the BMD or Z score of the femoral neck and baseline testosterone levels (Fig. 2C,D) and the BMD or Z score of Ward’s triangle and baseline testosterone levels (Fig. 2E,F) in all patients with KS.
In all, 33 men with KS had normal (>3.0 ng/mL) and seven had low serum testosterone levels (<3.0 ng/mL); the two groups were significantly different from the controls (Fig. 3). There was a statistically significant linear correlation between the baseline testosterone level and BMD or Z score of the spine (Fig. 4A,B), the BMD or Z score of Ward’s triangle (Fig. 4E,F) and a weak linear correlation between the BMD or Z-score of the femoral neck (Fig. 4C,D) in low-testosterone group with KS.
KS was first described in 1942 as an endocrine disorder characterized by small firm testes, gynaecomastia, hypogonadism, and higher than normal concentrations of FSH . KS is the commonest form of hypogonadism and of chromosome aneuploidy in males; ≈ 80% of cases are due to a congenital numerical chromosome aberration 47XXY; the remaining 20% have higher-grade chromosome aneuploidies, e.g. 48XXXY; 48XXYY; 49XXXXY, 46XY/47XXY mosaicism, or structurally abnormal X chromosomes [15,16].
The clinical presentation of patients who come to medical attention varies according to age. For patients presenting before puberty, only discrete physical anomalies such as a slightly lower than normal testicular volume or long extremities might be noticed. Sexual development can be normal before puberty and includes the initiation of normal pubertal changes and normal pituitary-gonadal function . In adolescence and after puberty the syndrome is characterized by small firm testes and varying symptoms of androgen deficiency. At the time of puberty characteristic skeletal proportions begin to develop. These patients are generally of average height or taller. Serum testosterone concentrations increase during early adolescence in some patients and then begin to decrease by 15 years old; it is lower than normal in ≈ 80% of adult patients with a 47XXY karyotype . The exact mechanism to explain the androgen deficiency is unknown, and the degree of Leydig cell dysfunction is variable. On average, the oestradiol concentration is higher than in normal men. Serum concentrations of sex-hormone binding globulin are high, causing a further decrease in biologically active free testosterone. Concentrations of LH and FSH are high in most cases. FSH shows the best discrimination, and has little overlap with values in men with normal karyotypes, and results in damage to the seminiferous tubules .
Several studies have explained the effects of sex hormones on bone cells. Osteoblasts carry α and β nuclear receptors for oestrogens , and androgen receptors have also been identified . In addition to this genomic pathway, a non-genomic pathway involves receptors to male or female sex steroids that contribute to the regulation of osteoblast apoptosis . Thus, in both men and women, a decrease in circulating free sex steroids (testosterone and oestrogens) leads to osteopenia or osteoporosis .
Horowitz et al. first published the relationship between BMD and serum testosterone levels in men with KS; they reported that serum dehydroepiandrosterone sulphate and testosterone were significantly related in men with KS (r = 0.64, P < 0.001), but not in controls (r = 0.22, not significant). Forearm BMD and fat-corrected forearm BMD were significantly related to serum testosterone levels in the KS group (r < 0.63; P < 0.01), but not in the control subjects (r < 0.03, not significant).
However, there is debate about whether testosterone replacement would increase the BMD in hypogonadal men. There are several studies reporting that sufficient testosterone replacement, with the currently available methods, does not reverse the decrease in bone mass associated with hypogonadism in patients with KS [24,25].
There is no study evaluating testosterone replacement in men with KS who have a low serum testosterone level. In the present study, serum testosterone level and BMD had a linear relationship in men with KS who also had a low testosterone level. This finding was especially notable for the BMD at the spine and femoral Ward’s triangle, where there was a statistically significant linear correlation between serum testosterone level and BMD. Therefore, testosterone replacement should be effective in this group of men with KS and a low serum testosterone level.
Recently several investigators suggested that oestrogens are more effective in restoring bone mass than are androgens. In a study of 315 healthy people, Khosla et al. found correlations between forearm BMD and levels of bioavailable oestradiol in subjects aged >60 years (radius, r = 0.29, P < 0.01; ulna, r = 0.33, P < 0.001). However, in middle-aged subjects the correlation was significant only in the ulna (r = 0.21, P < 0.05) and the correlation was weaker. This group found no significant relationship between total blood oestradiol and bone mass in middle-aged subjects. A second longitudinal study, published in summary form and on healthy elderly subjects (mean age 75.5 years) followed over a 4-year period, showed that free oestradiol and bioavailable oestradiol correlated weakly with the forearm BMD (r = 0.18, P = 0.02) .
However, the potential role of oestrogens in the pathophysiology of male osteoporosis has been discussed. Thus, in 1994, Smith et al. reported a case of a young patient with a mutation of the oestrogen-receptor gene associated with a decrease in bone density, and with normal testosterone levels. Morishima et al. and Carani et al. then described two patients with a mutation of the aromatase gene, who also had a decrease in bone density.
The present study has some limitations: First, due to the short follow-up, we do not have sufficient data to support the view that testosterone replacement is effective in restoring the BMD in men with KS. Second, relatively few patients were assessed; and finally, we measured only total testosterone and not the free form.
In conclusion, there was a relationship between a lower BMD and testosterone level in patients with non-mosaic KS, suggesting that low testosterone levels cause inadequate bone development and low a BMD in these patients. Therefore, testosterone replacement might be necessary to prevent bone mineral deficiency and future risk of fractures in men with KS who have a low serum testosterone level.