Hypogonadism is one of the most common causes of male osteoporosis. Testosterone and estrogen both play roles in bone metabolism. These hormones aid in bone formation and help prevent bone resorption [6, 7, 8,9]; interestingly, estrogen may play a bigger role in halting resorption than testosterone [8]. Indeed, 70% of gonadal effect on male BMD may be due to estrogen [8]. This contribution of estrogen is increasingly more recognized. Barrett-Connor et al. [10] found that a population of men with a known vertebral fracture had low estradiol levels even without reduced testosterone levels. Gennari et al. [11] followed 200 men over 4 years and demonstrated the decline in testosterone and bioavailable estradiol levels with age. The estradiol levels negatively correlated with changes in BMD and with bone turnover markers, which was not seen with testosterone levels. Therefore, although testosterone does play a role in achieving and sustaining BMD, estrogen is emerging as the more dominant factor in this capacity.

Besides aging, other conditions can produce a hypogonadal state. One such situation is androgen deprivation therapy (ADT) for the treatment of prostate carcinoma. A number of studies have been published that examined the relationship between this therapy and BMD decline. A small study by Mittan et al. [12] demonstrated that after 12 months of gonadotropin-releasing hormone (GnRH) therapy, a significant decrease in BMD at the hip and the ultra distal radius occurred compared with the control group. N-telopeptide (NTX) was increased at 6 and 12 months. Two other studies compared patients on GnRH agonist therapy with healthy patients and to patients with prostate cancer who were not on ADT. Both studies showed that BMD decreased significantly in those on therapy. NTX was again elevated, and alkaline phosphatase was increased in the Stoch et al. study [13]. Basaria et al. [14] also found a correlation between the length of therapy and the decrease in BMD.

More important than the decrease in BMD is the effect of ADT on fracture risk. Investigators have examined the risk of fractures in patients on
ADT. Shahinian et al. [15] assessed more than 50,000 men with prostate cancer and compared them with those who either had received ADT within the first 6 months after diagnosis or with those who had not received such therapy. In the first 5 years after diagnosis, those who were in the treatment group had significantly higher fracture rates than those who were not; 19.4% versus 12.6%, respectively. There was a positive correlation between fracture risk and the number of GnRH agonist doses. Those who had undergone orchiectomy had the highest risk of the treatment subgroups, with a relative risk for fracture of 1.54. In this study, the number needed to harm by ADT was 28 for any dose of a GnRH agonist and 16 for orchiectomy. Lopez and others [16] compared men who received LHRH agonists, with or without peripheral androgen receptor blockers, with controls over 4 to 5 years. They found that over this time, 11% of the treatment group and 4% of the control group fractured. The members of the treatment group were often older, more drank alcohol, and more had already sustained a fracture.

Hypogonadism, particularly hypoestrogenism, can be seen throughout the life cycle of women. In young women, diseases that cause low estrogen states result in decreased peak bone mass [2,5]. Two common examples are amenorrhea and anorexia nervosa (AN).

Amenorrhea can lead to osteoporosis even in young women. If a woman misses half of her menses by age 20, she can experience a decrease in BMD [5]. In addition to the severity of the disturbance in the menstrual cycle, the length of this alteration may also affect bone density [17]. In a study of amenorrheic and normal dancers and nondancers, the amenorrheic subjects had lower baseline BMD, up to 8% less than controls, and their BMDs remained lower during the 2 years of follow-up [18]. The amenorrheic subjects whose menses restarted during the study experienced an increase in BMD, but again it did not reach that of the controls. In addition, decreased spine BMD correlated with the development of stress fractures.

Anorexia is another common cause of amenorrhea in young women, and the resulting hormonal and nutritional deficiencies contribute to the decreased BMD in this situation [2,5]. Soyka et al. [17] showed that the majority of 130 anorexic women that they studied had decreased bone mineral density on dual-energy x-ray absorptiometry (DXA) scan, with more than 90% having T-scores of -1 or less and almost 40% with scores of -2.5 or below [17]. Weight, the age at first menses, and the duration of amenorrhea increased one’s risk for decreased BMD. In addition to the effects of hypoestrogenemia, these women can also have secondary hyperparathyroidism from calcium and vitamin D deficiency, hypercortisolemia, and malnourishment [2,5]. Insulin-like growth factor 1 (IGF-1) may also play a role.

Investigators have explored the effect of decreased IGF-1 concentrations in anorexic women. Soyka et al. [19] found significantly decreased IGF-1 levels in those with AN compared with controls, as well as a correlation between IGF-1 levels and such reflections of nutritional status as body mass index, percent body fat, lean body mass, and leptin, but not levels of estradiol [19].

What to do when decreased BMD is found among anorexics is a topic of debate. Whether successful therapy of anorexia improves BMD is controversial [5]. A 2002 study did not demonstrate a significant increase in BMD among subjects with anorexia who gained weight after a 1-year follow-up [20]. Whether oral contraceptives (OCP) help in this setting was studied in a trial that compared 65 women with AN to 52 healthy controls [21]. Of those with anorexia, 16 had who had taken OCPs were found to have significantly higher lumbar spine BMD, although it was still lower than the BMD of the control group [21]. However, subsequent longitudinal trials did not show an increase in BMD with OCPs or hormone replacement therapy (HRT) alone [22,23].

One must remember that not all anorexics are women. Up to 5% to 10% of patients with AN are male [2]. Although osteoporosis has been reported in male anorexics, limited data are available regarding its prevalence and the effects of treatment. Further studies need to be conducted regarding this issue.

Immunosuppressants, such as cyclosporine and tacrolimus, have been shown to exert negative effects on bone density in animal models [41,42], but these effects in clinical trials have been less clear. Although decreased BMD has been seen in patients taking these medications, a possible confounder may be the concomitant use of glucocorticoids in these cases [43].
Their effect on fracture risk is also uncertain; for instance, one trial did not find these medications to be significantly correlated [44].

Hyperthyroidism can cause osteoporosis as well, decreasing BMD by 10% to 30% in women [5]. Both osteoclastic and osteoblastic activities are increased in hyperthyroidism; however, osteoclasts are amplified more [5]. Hyperthyroidism may also decrease intestinal calcium absorption [2].

The cause of the hyperthyroidism does not appear to affect the degree of bone loss. Jodar et al. [45] compared the BMDs of overtly hyperthyroid, controlled hyperthyroid, and healthy patients. They showed a lower BMD in the overtly hyperthyroid and the controlled hyperthyroid patients compared with the healthy ones. Overtly hyperthyroid patients had more bone loss than the controlled ones. No significant difference in bone density between Graves disease patients and those with toxic multinodular goiter was found. Obtaining a euthyroid state is important for increasing BMD. Kumeda et al. [46] demonstrated that patients with hyperthyroidism on drug therapy, who still had a low thyroid-stimulating hormone (TSH) despite normal thyroid hormone levels, had persistent increases in bone turnover markers. With resolution of thyroid disease, BMD increases but may not normalize [5]. Some data supporting normalization of bone density to age-expected BMD do exist, however. For instance, Karga et al. [47] compared both untreated and treated hyperthyroid patients with age-matched women without prior hyperthyroidism. They found that although BMD was lower than that of controls in untreated hyperthyroidism for up to 3 years after treatment, BMD was not significantly different from that in controls at least 3 years after therapy.

Decreased vitamin D concentrations can have deleterious effects on bone metabolism. Two important metabolic bone derangements that can result are secondary hyperparathyroidism and osteomalacia. Secondary hyperparathyroidism develops when the vitamin D concentration decreases to an insufficient and deficient level [47]. Consequently, one can see increased bone turnover [48,49] and decreased BMD [49, 50,51]. Lips [48] suggested that vitamin D insufficiency ranges from 10 to 20 ng/mL (25–50 nmol/L). Variations do exist, as Chapuy [52] found that parathyroid hormone (PTH) increased at a higher 25(OH)D threshold, 31 ng/mL (78 nmol/L). Osteomalacia may develop after prolonged vitamin D deficiency [48,50]. According to Lips [48], this can be found at 25(OH)D levels of less than 5 ng/mL (<12.5 nmol/L). Various trials have further substantiated these theories, confirming an inverse correlation between PTH and 25(OH)D [4,48,52]. In addition, the presence of increased bone turnover can be inferred by the inverse relationship between bone turnover markers and vitamin D levels [49,53]. Interestingly, a study reported a stronger inverse correlation between 24-hour urinary calcium excretion and PTH than 25-OHD and PTH level [4].

The gold standard in diagnosing osteomalacia is bone biopsy. This is not always readily available, however. Frequently, the diagnosis can be inferred from the presence of high total or bone-specific alkaline phosphatase, in the
setting of a low vitamin D level, hypocalciuria, and secondary hyperparathyroidism. These individuals complain of bone pain (deep aching hip or pelvic pain), motor weakness, and recurrent fractures.

The loss of bone density among vitamin D–deficient individuals has been demonstrated in various trials. For instance, Mezquita-Raya et al. [51] showed a positive correlation between 25(OH)D concentrations and BMD. Elsewhere, a 4% lower trochanter BMD has been demonstrated with 25(OH)D levels less than 10 ng/mL (25 nmol/L) [49]. In addition, Sahota et al. [53] reported a significant decrease in BMD at the hip with 25(OH)D concentrations between 6.1 and 12 ng/mL (15.25 to 30 nmol/L) [53]. One might further reason that decreased vitamin D is linked to an increase in risk of fracture. These data are not as firm, but there have been reports of an association [48,54].

The effect of this deficiency on bone density can also be illustrated by changes that occur after vitamin D is supplemented. With such replacement, PTH concentration decreases, bone turnover calms, and BMD improves [48,55]. In addition, the incidence of fractures decreases. For instance, Adams et al. [55] showed a 4% to 5% increase in BMD at the lumbar spine and the femoral neck per year with vitamin D repletion. Also, Dawson-Hughes et al. [56] demonstrated a significant increase in BMD and a significant decrease in nonvertebral fractures with calcium and vitamin D supplementation over 3 years. A metaanalysis showed that vitamin D reduced the incidence of vertebral fractures [relative risk (RR) 0.63, 95% confidence interval (CI) 0.45–0.88, p < 0.01] and showed a trend toward reduced incidence of nonvertebral fractures (RR 0.77, 95% CI 0.57–1.04, p = 0.09) [57].

Vitamin D deficiency is increasingly being recognized as a contributing factor to osteoporosis. In fact, Favus [58] has postulated that it is possibly the most common etiology [58]. Numerous studies assessing its prevalence have been performed. The Multiple Outcomes of Raloxifene Evaluation clinical trial found 4% of their subjects with a 25(OH)D of less than 10 ng/mL (<25 nmol/L) and almost one-quarter with a level of 10 to 20 ng/mL (25–50 nmol/L) [49]. Holick et al. [59] found that half of the women that he evaluated, postmenopausal women in North America on osteoporosis treatment, had a 25(OH)D level of less than 30 ng/mL (75 nmol/L). Chapuy et al. [52] saw a 25(OH)D concentration of 12 ng/mL (30 nmol/L) or less in 14% of the French women they studied. In addition, 75% of their subjects had a 25(OH)D level of less than 31 ng/mL (78 nmol/L). In a study of 104 patients at least 98 years old, 99 had an undetectable concentration of 25(OH)D [60]. Although this deficiency can be seen among ambulatory patients, it is more common in hospitalized patients, those living in nursing homes, and patients who have sustained hip fractures [48,54]. For instance, a study of hospitalized general medicine patients revealed that 57% had a 25(OH)D level of less than 15 ng/mL (37.5 nmol/L). Seventy-seven of the patients in this study were younger than 65 years old and were without known risks for vitamin D deficiency; 42% of this subgroup was deficient [61]. Although not as common, decreased concentrations of vitamin D are also seen in the younger population. For example, Tangpricha et al. [62] found that 36% of the patients aged 18 to 29 years in their study had 25(OH)D levels less than
20 ng/mL (50 nmol/L) at the end of winter; this decreased to less than 10% at summer’s end.

Several reasons exist for the increased prevalence of vitamin D deficiency among the elderly. These include less sun exposure [48,54,58,63], decreased ability to synthesize vitamin D₃ [48,58,63], inadequate vitamin D intake [48,52,58,63], and decreased intestinal responsiveness to vitamin D [64]. In addition, dysfunction of the liver or the kidney can contribute to the deficiency [54,58], as can the ingestion of medications that increase vitamin D clearance [58].

Since much of one’s vitamin D supply comes from sun exposure, it seems logical that concentrations may fluctuate with the seasons and with location. Multiple studies have confirmed this seasonal variation, with lower levels in the winter than in the summer [49,62,65]. This has also been shown with respect to the inverse relationships of 25(OH)D with PTH [48,54,62] and with bone turnover markers [48]. In addition, Chapuy et al. [52] found a location difference in 25(OH)D concentrations, with a significant decrease in the north in comparison to the south of France.

Primary hyperparathyroidism can cause decreased BMD and increased risk of fractures. This occurs mainly due to an increase in bone resorption [50]. Parathyroid hormone is likely catabolic in cortical bone, such as in the distal radius, and anabolic in cancellous bone, such as in the lumbar spine [66, 67,68]. Thus, usually more cortical bone is lost than cancellous bone [51,68], although decreased density of cancellous bone has been seen [66, 67,68]. The correlation between parathyroid hormone and BMD has been shown in such studies as that by Sitges-Serra et al. [69], who found a significantly increased PTH concentration in osteoporotic patients. Although it seems that this decreased bone density would correlate with an increase in fractures, results are not consistent. Khosla et al. [70] compared the fracture risk in more than 400 patients with mild primary hyperparathyroidism to that in the general population. They found a significant risk of vertebral and pelvic fractures. Other trials have not found an increased risk of fractures, however [66,67].

Effects of primary hyperparathyroidism can also be seen by assessing bone density and fractures after parathyroidectomy. Several studies have investigated these changes, and they have demonstrated an increased bone mineral density and a decreased fracture risk after surgery. A study by Silverberg et al. [71] followed 121 patients with primary hyperparathyroidism for 10 years. They noted that the patients who underwent parathyroidectomy experienced an increase in BMD—8% at the lumbar spine and 6% at the femoral neck after the first year, then 12% at the lumbar spine and 14% at the femoral neck after 10 years. Sitges-Serra et al. [69] compared the DXA scans of 28 hyperparathyroid patients taken before and 1 year after parathyroidectomy. They demonstrated a significant increase in BMD at the femoral neck and the proximal femur, especially in the osteoporotic patients. Rao et al. [72] studied 53 patients with mild primary hyperparathyroidism and randomized these subjects to surgery or no surgery. Small but significant yearly increases in BMD were seen at the femoral neck and at the total hip after parathyroidectomy.
Finally, Vestergaard and Mosekilde [73] showed a 31% decreased risk of fracture after parathyroidectomy.

Multiple factors contribute to the decline in BMD in patients with solid organ failure, as well as those who have received transplants. The disease itself, risk factors that develop because of the disease, and medications, including loop diuretics and glucocorticoids, all contribute to the decreased BMD. Many of the patients with end-organ failure are older, immobile, malnourished, Caucasian, vitamin D–deficient, hypogonadal, tobacco users, or alcohol consumers [5,42,74]. After transplantation, many of these risks still exist. In addition, of prime importance are the immunosuppressants that are used in the post-transplant period [5,42].

Osteoporosis has been documented in these patients before and after transplant. In pretransplant patients, it has been reported in up to 19% of those with heart failure [74], 29% to 61% with lung disease [42,75,76], and 26% to 52% with liver failure [42,77]. Fractures have also been reported in these patients; for instance, 25% to 29% of end-stage lung disease patients were found to have suffered a vertebral fracture in one study [75]. After transplant, much of the density loss occurs in the first several months [5,42]. Shane et al. [78] showed that cardiac transplant patients lost a mean of 7% to 11% of the density in their lumbar spine and femoral necks in the first year; most of this occurred in the first 6 months in the lumbar spine [78]. This is similar to other reported data [42]. Postliver transplant patients may lose up to one-quarter of their BMD in the spine in the first year, although this is controversial [42]. Bone density in liver transplant patients may improve as early as 6 months after transplant, at times increasing to pretransplant density [5,42,79,80]. After lung transplant, 2% to 5% of bone density is lost in the first year [42]. An increased risk of fracture after transplant has also been reported. In the first year, up to two-thirds of patients can fracture [81]. Leidig-Bruckner et al. [82] found that after two years, about one-fourth of the heart transplant patients and about one-fifth of the liver transplant patients suffered a vertebral fracture [82].