Disorders Associated With Acute Rapid and Severe Bone Loss


  • Solomon Epstein,

    Corresponding author
    1. Mount Sinai Bone Program, Departments of Medicine, Geriatrics and Physiology, Mount Sinai School of Medicine, Division of Endocrinology, Department of Medicine, Bronx VA Medical Center, New York, New York, USA
    • Address reprint requests to: Solomon Epstein, MD Mount Sinai Bone Program Endocrinology (1055) 1 Gustave L. Levy Place New York, NY 10029, USA
    Search for more papers by this author
  • Angela M Inzerillo,

    1. Mount Sinai Bone Program, Departments of Medicine, Geriatrics and Physiology, Mount Sinai School of Medicine, Division of Endocrinology, Department of Medicine, Bronx VA Medical Center, New York, New York, USA
    Search for more papers by this author
  • John Caminis,

    1. Arthritis/Bone Section, Novartis Pharmaceutical Corp., East Hanover, New Jersey, USA
    Search for more papers by this author
  • Mone Zaidi

    1. Mount Sinai Bone Program, Departments of Medicine, Geriatrics and Physiology, Mount Sinai School of Medicine, Division of Endocrinology, Department of Medicine, Bronx VA Medical Center, New York, New York, USA
    Search for more papers by this author

  • Dr Inzerillo has served as a consultant for Novartis Pharmaceuticals and has received a grant from ARA/Pfizer Inc. Dr Caminis is an employee of Novartis Pharmaceuticals. Dr Epstein has served as a consultant for Merck & Co., Inc., Novartis Pharmaceuticals, NPS Pharmaceuticals, Roche, and Wyeth. Dr Zaidi has received a grant from Procter & Gamble.


We describe a constellation of bone diseases characterized by the common feature of acute, rapid, and severe bone loss accompanied by dramatic fracture rates. These disorders are poorly recognized, resulting mainly from systemic diseases, frailty, immobilization, and immunosuppressive drugs, such as glucocorticoids and the calcineurin inhibitors. The opportunity to prevent or treat fractures is commonly missed because they are often not detected. Ideally, patients need to be identified early and preventative therapy initiated promptly to avoid the rapid bone loss and fractures. The most effective therapy at present seems to be the bisphosphonates, particularly when bone resorption is predominant. However, more severe forms of bone loss that result from an osteoblastic defect and reduced bone formation may benefit potentially more from newer anabolic agents, such as recombinant human parathyroid hormone (rhPTH).


This review describes a spectrum of bone diseases characterized by the common feature of acute, rapid, and severe bone loss (ARSBL) accompanied by a high fracture rate often far greater than that seen with postmenopausal and age-related osteoporosis. These disorders consist mainly of systemic diseases associated with drug therapy, frailty, immobilization, and the impact of the systemic disease itself. Examples, which will form the basis of our discussion, include glucocorticoid-induced osteoporosis (GIOP), post-transplantation bone disease, stroke, and immobilization from an active state (e.g., after acute spinal cord injury). We have chosen not to include rapid bone loss after estrogen therapy withdrawal(1,2) or patients initiating treatment with depot progesterone compounds(3,4) because there seems to be better recognition of these disorders, and data on acute fracture incidence are lacking.

The average annual bone loss reported in these disorders is approximately 8% at the spine and 5% at the hip. This corresponds to what is seen in “fast” bone loss in postmenopausal osteoporosis, but the definition is arbitrary and the rate depends on the etiology. In the most severe forms of microgravity-induced bone loss, the bone loss in 1 month, approximately 2%, equals that of a postmenopausal woman in 1 year.(5) In post-transplant osteoporosis, astounding fracture rates as high as 65% per year have been reported.(6) Clinically, therefore, it is important that we recognize this constellation of disorders, its potentially devastating effects on the skeleton, and the rapidity at which the bone loss occurs. Effective management can prevent, or in instances, reverse the ensuing bone loss.

Pathophysiologically, diseases associated with osteoporosis are associated with an imbalance of bone remodeling, a process by which new bone, laid down by the osteoblast, replaces the old bone that is removed by the osteoclast. In the initial phase, the rapid bone loss may be in part caused by increased osteoclastic activity, the mechanism of which is unclear, although pathophysiological factors have been recognized (e.g., estrogen deficiency). This is followed by a phase of predominant reduction in osteoblastic bone formation (except with calcineurin inhibitor [CI]-induced bone loss).(7–10) Additionally, other factors influence the bone remodeling unit in these disorders. These include immobilization and bone quality,(11) with fractures possibly occurring at a higher bone mineral density (BMD).(12) An example illustrating some of these features may be related to the impact of glucocorticoids at all levels in terms of bone mineralization, transcription of collagenases, osteocalcin synthesis, and growth factor protection. The increased osteocyte apoptosis, which may reduce the response of the osteocyte functioning as a mechanosensor leading to impaired ability to resist load with increased susceptibility to microcrack damage and hence fracture, all contribute to the deleterious effects of glucocorticoids.(7,8,13)


This is probably the most common form of secondary osteoporosis.(14) The use of glucocorticoids in the general population has been estimated to vary between 0.5% and 1.7% in postmenopausal women over 55 years of age.(15) The incidence of GIOP is approximately 50% in all patients treated for 6 months or longer.(16,17) The estimated fracture incidence, predominantly at the spine and ribs, in patients on long-term glucocorticoids is 34%.(18) The risk of hip fracture is, however, doubled in glucocorticoid users.(19) Readers can refer to comprehensive reviews for more details.(20–23)

It is not only the dose but also the duration of treatment that determines GIOP. Daily oral doses between 2.5 and 7.5 mg/day can increase fracture risk.(17) Inhaled glucocorticoids are not innocuous, and recent studies attest to their deleterious effects on bone density.(24) The risk of fracture seems to be at the maximum within 6 months of commencing therapy but does decrease after therapy is stopped.(17) However, if patients have low bone mass to begin with, even after stopping therapy, fracture risk may still remain high, particularly in postmenopausal females without estrogen replacement therapy.(25) Recently an increased incidence of fracture has been reported in children exposed to oral corticosteroids, and because this group is now undergoing organ transplantation frequently, this will become a serious concern.(26)

Bone loss in GIOP occurs in both trabecular and cortical bone. Trabecular bone seems to be the most susceptible, initially with an increased rate of vertebral fractures. However, it remains unclear whether fractures occur at a lower threshold (i.e., T-score) than that seen with involutional osteoporosis.(27) This invokes the concept that bone quality (including microarchitecture), as opposed to bone density alone, is also important in determining fracture risk in these patients.(28)

There seems to be two phases contributing to the bone loss associated with GIOP. The first phase is characterized by increased bone resorption. The second phase (i.e., decreased bone formation) is probably the predominant defect. This etiology of GIOP has been subject to intense research at the molecular, cellular, and tissue levels; this has been reviewed extensively.(29–34) It should be noted that there are no ideal animal models of GIOP to show these changes in vivo.(35,36) Figure 1 shows a schematic representation of the multitude of mechanisms contributing to cellular dysfunction. There is compelling evidence that the osteoblast bears the brunt of the insult from the glucocorticoid excess. Inhibition of its mineralizing and osteoclast-support function has been documented in in vitro studies.(7) Thus, glucocorticoids have been shown to inhibit the expression of a variety of key osteoblastic genes, including the collagen A1, fibronectin, transforming growth factor β (TGFβ), insulin-like growth factor-1 (IGF-1), and RANKL genes.(9,37–43) Specifically, the reduction in the expression of RANKL, a cytokine that promotes osteoclast formation, likely results in low osteoclast numbers and reduced bone resorption that ensues.(44,45) Taken together, the lowered bone resorption and formation levels can ultimately result in low-turnover bone loss. In addition, direct effects of glucocorticoids in decreasing the life span of osteoblasts and hence their ability to lay down new bone through accelerated apoptosis (or programmed cell death) has been documented and contributes to the deleterious effect on bone mass and fracture risk.(7)

Figure FIG. 1..

Glucocorticoids have temporal-dependent effects on bone. By reducing the production of integrin receptors, fibronectin, and collagen, they initially inhibit bone formation (arrows, right side). Glucocorticoids also initially induce bone resorption by increasing pro-osteoclastic factors, such as RANKL. Over time, osteoblast numbers continue to decrease with osteoclast numbers following; this may be because of either decreased proliferation or increased apoptosis. GR, glucocorticoid receptor.

The pathophysiology of the high remodeling seen in the initial stages of GIOP remains unclear. This bone loss, which is often transient, can be explained not only by a decrease in sex steroids,(46,47) but through indirect effects on parathyroid hormone (PTH) secretion.(14) Glucocorticoid-induced hypercalciuria and decreased calcium absorption can potentially lead to secondary hyperparathyroidism.(48,49) However, the contribution of secondary hyperparathyroidism to GIOP is controversial, because most studies do not show elevated PTH levels above the normal values, and the bone histomorphometry does not support this concept.(50) An increased sensitivity at a receptor level to circulating PTH levels, however, cannot be excluded. To these effects, the added burden of malnutrition, immobility, and the underlying disease for which glucocorticoids are prescribed need to be added.


This form of osteoporosis has emerged mainly because of the advent of potent immunosuppressive agents. While drugs such as CIs have increased longevity by decreasing rejection episodes they do so at the cost of and its resultant fractures. The number of new transplants in the United States is in the region of 20,000 per year.(51) Over one-quarter million patients have already undergone solid organ transplant, and it is unfortunate that without the realization of the magnitude and rapidity of their bone loss, they remain at a high risk of fracture.

Mechanistic insights

Post-transplant bone loss is multifactorial; immobilization, poor nutrition, hypogonadism, and the underlying disease all play a role, although the contribution of each is unclear. However, the main culprits are the immunosuppressants glucocorticoids and CIs; the latter include, most notably, cyclosporine (CsA) and tacrolimus (FK506). Glucocorticoids probably are the most deleterious to bone initially when large doses are used soon after transplantation. Even in the absence of glucocorticoids, CIs can cause profound bone loss.(52,53)

What is the mechanism of CI-induced bone loss (Table 1)? Originally, the deleterious effect of CsA was demonstrated in rats where severe and rapid bone loss occurred with a metabolic profile that revealed increased osteocalcin and 1,25-dihydroxyvitamin D synthesis.(54–59) PTH levels were largely normal. Bone histomorphometry showed a high-turnover bone remodeling state with both resorption and formation increased, but with resorption far exceeding bone formation.(55) This also occurs with FK506 (unlike GIOP where the turnover is much less pronounced and formation is inhibited to a greater extent than resorption(59)). This same picture has been seen in clinical practice in bone biopsy specimens from organ recipients while on immunosuppressant therapy.(60) Newer immunosuppressants, including rapamycin(61) and mycophenolate mofetil,(62) when administered to rats in vivo, do not cause bone loss. They therefore have promise in allowing lower doses of bone-active immunosuppressants to be used.(63) Azathioprine, which is commonly used as a third immunosuppressant in combination with glucocorticoids and CIs, does not affect bone in short-term rat experiments.(64)

Table Table 1. Action of Cyclosporine A on Bone In Vivo
original image

Further insights into the mechanism of the CI's have demonstrated that the T-lymphocyte is critical for the action of CsA, and presumably FK506, to cause bone loss in vivo.(65) The exact mechanism is incompletely understood, and the role of which specific subset of T-lymphocytes as well as the role of the B-lymphocyte in the process are currently unknown. An attractive hypothesis may involve the effect of CIs on the osteoprotegerin (OPG)-OPG ligand (OPGL) system, because the T-lymphocyte is a source of these factors.(66) This needs to be further explored. The picture is complicated because in vitro CsA and FK506 both cause an inhibition of bone resorption in the isolated osteoclast system.(67) They also similarly inhibit osteoclast formation in bone marrow cultures.(68) The target for both CsA and FK506 is calcineurin, a cellular phosphatase that is potently inhibited by both drugs.(69–71) Several calcineurin genes have now been cloned and sequenced. Both calcineurin Aα and Aβ isoforms are expressed in osteoblasts and osteoclasts.(72) The calcineurin Aα knockout mouse, as would be expected, exhibits an osteoporotic phenotype.(73) However, the osteoporosis, contrary to what one would expect from the effects of CsA and FK506 in vivo, was of the low-turnover variety. The overall interpretation may be that drug effects on the T-lymphocyte, which then produces osteoclast stimulatory cytokines, overwhelm the effects of CIs on calcineurin in bone. However, blocking the bone-resorbing cytokine TNFα did not prevent the CsA-induced bone loss in vivo.(73)

Clinical syndromes produced by organ transplantation

It should be realized that the discussion focuses on acute bone loss, which decreases after the initial early phase but may continue albeit at a slower rate. The presentation is usually symptomatic spinal fracture but most are asymptomatic and detected incidentally. Hip fractures are by their nature always symptomatic, presenting spontaneously or after a fall or eccentric rotation of the femur. Before, transplantation fractures occurred because of the severity of the underlying disease, the time waiting for transplantation, the drugs used, anorexia, and immobilization. Post-organ transplantation, the rate of bone loss and the different incidence of fracture are dependent on the organ transplanted and the underlying disease as well as the dose of immunosuppressants used to prevent rejection. Different immunosuppressant doses and schedules are used by different centers. Thus, they can vary from center to center, and this probably accounts for the wide range of fracture incidence reported in different published reports. Below, we will discuss the relevant clinical features and distinctive presentations of kidney, kidney-pancreas, cardiac, liver, lung, and bone marrow transplant. It is highly unlikely that patients undergoing transplant of other organs, such the intestines, will present differently, but there is no bone density or fracture data currently on any of the newer varieties of organ transplant.

Kidney and kidney-pancreas transplantation:

The rate of bone loss and fracture after kidney transplant is less than that of other organs. Reasons for this may be the long experience that renal physicians have with transplantation and the low dose of immunosuppressants used. However, renal transplant patients almost always have underlying bone disease, particularly if they have been on long-term dialysis. The spectrum of renal osteodystrophy includes osteoporosis, osteomalacia, secondary hyperparathyroidism, sclerosteosis, adynamic bone disease, and mixed bone disease. After transplantation, most abnormalities correct, but persistent hyperparathyroidism, hypercalcemia, adynamic bone disease, and avascular necrosis may be difficult to manage.

The rate of bone loss after kidney transplant varies and is between 6% and 18% per year at the spine(74) and ∼4% per year at the hip.(75) Bone loss occurs mostly within the first 6 months after transplant.(74) Interestingly, males lose more bone in the proximal femur than females.(76,77) Spinal BMD recovers with a decrease in immunosuppressant dosage, but hip BMD generally continues to decline. Overall, rates of fracture vary between 10% and 20% in the first year.(78) However, in diabetic patients undergoing a single or double transplant, the incidence of fracture, particularly of small bones, can approach a staggering 50% within 1–2 years.(79–81) The reason for this high rate and distribution of fracture in diabetics is unknown, except that adynamic bone disease is more frequent in these patients and may indeed play a role. Other factors, including microvascular disease, neuropathy, and the deleterious effect of hyperglycemia on the osteoblast, may also contribute.

Histomorphometric data on bone biopsy specimens are scant. It is possible, however, that the early effects of glucocorticoids at 6 months may be superimposed by the high-turnover remodeling state induced by CIs that, in experimental models, is not unlike secondary hyperparathyroidism. Nonetheless, current clinical trial evidence indicates that the extent and rate of bone loss in patients receiving either glucocorticoids or cyclosporine is equivalent.(82) Whether this will translate into similar fracture rates is unknown.

Cardiac transplantation:

The form of osteoporosis post-cardiac transplant is one that produces much morbidity with fracture risk in cross-sectional studies varying between 18% and 50% at the spine.(78,83–94) By the time of transplantation, most patients have lost substantial bone density at both the spine and femur. When applying World Health Organization (WHO) criterion for osteoporosis, 8–10% patients suffer from osteoporosis, and 40–50% have osteopenia.(83,84) Risk factors are similar to the other transplant groups, except for the osteodystrophy in renal transplant patients, and include lifestyle excesses of tobacco and alcohol, loop diuretics, inactivity, hypogonadism, and anorexia. The rate of bone loss at the spine is maximal in the first 6 months after transplant, coinciding with the greatest fracture incidence.(95) The decreases reported in lumbar spine BMD seem to taper off after 6 months. Experience is limited, but decreases in femoral neck BMD may also taper off with time. Despite traditional standard of care with calcium and vitamin D, fracture rates are high, with 36% patients suffering more than one fracture.(96) Both the pretransplant bone density and the rate of bone loss after transplantation seem to affect fracture incidence, but a normal baseline bone density does not preclude fracture after transplantation.(97)

Histomorphometric evaluation of bone biopsy specimens reveals a high turnover state accompanied by increased levels of serum osteocalcin and urinary markers of resorption.(85) This is again similar to that seen with CIs in experimental models, although secondary hyperparathyroidism caused by renal impairment cannot be excluded, even in the face of minimally elevated serum creatinine and PTH.

Liver transplantation:

The same underlying factors as with other organ transplants apply to liver transplantation except that the type of liver disease for which the patient is transplanted seems to have a profound influence on the rate of fractures after transplantation.(98–100) Patients with primary biliary cirrhosis have a high, ∼65%, incidence of atraumatic fractures of the spine, hip, ribs, and long bones within the first year after transplantation.(101,102) In general, the dosages of immunosuppressants used in liver transplant patients are higher than those used in renal and even cardiac patients. In addition, impaired liver function may impede cyclosporine metabolism, resulting in higher circulating levels. In both alcoholic and primary biliary cirrhosis, the bone disease is of a low turnover variety, which after transplantation, rapidly converts to a high remodeling state with an increase in markers of bone formation and resorption.(103,104) After transplant, bone density of the spine recovers in the first year,(94,105) but this has not been a uniform finding. In the femur, however, bone loss seems to continue despite lower glucocorticoid dosages.(106)

Lung transplantation:

These patients also have an extremely high prevalence of osteoporosis and osteopenia before transplantation in the range of ∼60% at the spine and 78% at the hip.(107,108) Vertebral fracture prevalence before transplantation varies between 25% and 29%,(107,109) and in cystic fibrosis, the reported fracture rate of ribs and vertebrae is increased by 10- and 100-fold, respectively.(108,109) Large immunosuppressant dosages post-transplantation result in a high, ∼37%, fracture incidence,(110) despite calcium and vitamin D supplementation and even occasional bisphosphonate therapy. The loss, mainly in trabecular bone, occurs within the first 6 months after transplant. The histological picture is once again of a high remodeling variety, with increased markers of both resorption and formation.(111) Note that high doses of glucocorticoids also impair bone growth in children undergoing such transplants.(112) It is unclear if CIs do the same, but in vivo, no abnormality in longitudinal growth was observed.(55) However, in a recent study, a mouse model, in which calcineurin Aα, the target for CIs, is knocked out, displays runting of long bones.(73) Whether this translates to the human situation is unclear.

Bone marrow transplantation:

The indications and frequency for bone marrow transplant is expanding rapidly. These patients suffer bone loss not so much from immunosuppressant therapy but from the myeloablative regimens that result in hypogonadism.(113–115) In women, this can be remedied by hormone replacement therapy.(116) Like all other transplant patients, however, low BMD may present before transplantation. Patients undergoing allogeneic transplants may be more at risk for bone loss at the femoral neck and spine mainly because of an increase in the incidence of graft-versus-host disease.(115) No fracture rates are available for bone marrow recipients.


Clinically, acute, and later chronic, immobilization results from among other factors, paralysis caused by trauma, poliomyelitis, multiple sclerosis, and cerebrovascular accidents. Prolonged immobilization because of fracture itself can cause substantial bone loss. Finally, a whole new emerging field achieving prominence relates to the effect of microgravity during space flight.(117) However, no fracture data or randomized controlled trials to show the effect of treatment are available.

Here the major factor is that of disuse with weakness and loss of muscle, resulting in a reduction of stresses and strains. The turkey ulna model represents the best model of disuse atrophy, which shows a dramatic loss of cortical bone within weeks of immobilization.(118) The widely accepted mechanism is that a reduction of compressive mechanical forces during immobilization results in reduced canalicular fluid flow in bone.(119) This causes hypoxia of osteocytes, the cells that transduce mechanical stimulation, and which in turn stimulate osteoclastic activity. Specific candidate molecules implicated in transducing mechanical stimuli into cellular signals include MAP kinases, the α subunit of the Na+ channel, and annexin V; none of these have however been unequivocally established as mechanotransducers.(120)

Bone loss caused by immobilization is acute, severe, and rapid. Histomorphometry in vivo shows reduced bone formation and transitory increases in bone resorption.(121) The net result is a low turnover state, which cannot tolerate or respond to normal stresses and strains. The bone loss maybe localized or generalized. It is localized to the affected side or limb when related to spinal cord injury or poliomyelitis. Most of the bone loss occurs during the first year after spinal cord injury but may continue up to 15 years after the event.(122) Trabecular bone is affected more severely and extensively than cortical bone. In paraplegic or tetraplegic patients, the rate of bone loss at the hip is ∼2% per month for the first 6 months and thereafter declines to 1% per month over the following 6 months.(123) The total bone loss can thus be substantial, up to 12%, over the first year. Tibial bone loss is even more severe-up to twice that at the femur.(124) Lumbar spine density is not decreased, likely because of adequate mechanical loading in the upright position.

Fracture data in spinal cord injury are mainly limited to the long bones. Fracture rates vary widely between 2.3% and 29.6%.(122) There are significant problems with interpreting these data, such as distinguishing fractures occurring at the time of injury versus those after injury and differentiating whether the reported fracture incidence refers to the number of fracture in the study population versus the actual number of patients with fractures. This reporting can lead to different interpretations as the same patient may suffer multiple fractures.


Stroke patients have demineralization and muscle atrophy,(125) which affects the paralyzed side; therefore, bone loss can be 9% after 1 year in those patients who do not relearn to walk.(126) The nonparetic leg may also be affected in those who do not learn to walk again, and bone loss can be ∼3% at the end of year 1. Fracture incidence is increased after stroke: in a study of 1139 patients admitted for stroke, the fracture incidence was between two and four times greater than a reference population. Most of the fractures (62.5%) affected the paretic side, and as expected, were mainly caused by falls.(127) Vitamin D deficiency is a factor in enhancing fracture risk after paresis. In stroke patients with vitamin D levels below 12 ng/ml, risk of fracture is significantly increased.(128)

A study examining hip fracture risk on a Swedish registry of patients with stroke found that, after hospitalization for stroke, the absolute risk of subsequent fracture within 1 year of hospitalization compared with that of the general population was increased by 7-fold.(129) The authors suggest that such patients should be preferentially targeted for treatment, and even short courses of therapy at the time of the stroke may provide substantial benefit to bone loss and fracture risk.


The clinical events of the underlying disease should alert the physician to the risk of bone loss and subsequent fracture. In these diseases, there is a need to measure BMD because this provides an assessment of fracture risk as well as a requirement to intervene to monitor the course of the disease and response to therapy. It is strongly recommended that BMD should be measured in patients awaiting transplantation, which can be years, so that treatment can be instituted even before transplant. After transplantation, BMD should be measured after 6 months, 1 year, and 2 years. For glucocorticoid administration, the same schedule is recommended; the first BMD measurement should coincide with the onset of therapy. For the other diseases, it is suggested that BMD be measured at the time of the event or as soon as possible after. DXA performed on the same machine and by the same technician, if possible, is the technique of choice. The sites should include both hips (if possible) and the lumbar spine. Overall, because of the rapid and severe bone loss, criteria for instituting therapy must be more stringent than those used for postmenopausal and involutional osteoporosis, and physicians should be more proactive with the use of effective antiresorptive therapy.

Markers of bone turnover may have an adjunct role in monitoring therapy of these patients, although markers of resorption may be more informative than the markers of formation (with GIOP as an exception) and include urine and serum cross-linked telopeptides. There is little data to substantiate their values for predicting or assessing their response to therapy either before or after organ transplantation. This area may also benefit from well-controlled prospective studies to provide useful information. A baseline value may be of help in assessing future response to treatment. Measurements ideally could be repeated at 3 months after treatment is initiated to assess a therapeutic response, as well as compliance. The test can then be repeated at yearly intervals, if necessary, although BMD and especially the lack of new fractures will be the most important yardstick of a therapeutic response.

Other useful biochemical tests include serum calcium and serum 25-hydroxyvitamin D (to exclude vitamin D deficiency). Measurement of sex steroids to exclude hypogonadism is prudent because some patients may benefit from replacement therapy. Interestingly, testosterone levels in males are low initially and tend to increase around 6 months after transplant, coinciding with a reduced dosage of immunosuppressants.(95,130) Testosterone treatment should therefore be stopped, and the patient should be reassessed for the continued need for testosterone replacement.

Twenty-four-hour urinary calcium levels may provide useful insights regarding the patient's calcium balance, specifically altered in malabsorption syndromes (e.g., in liver disease or intestinal disease). High urinary calcium levels of over 400 mg/24 h may reflect a tubular leak from CIs, such as tacrolimus and cyclosporine. Alternatively, it may arise from glucocorticoid treatment or excess vitamin D and calcium replacement.


There are a number of therapies that have been used in an attempt to prevent bone loss and fractures. However, because of the nature of the underlying diseases and the various doses of immune-modifying drugs used, study designs and outcomes have varied between studies. As an example, a randomized double-blinded prospective trial from one center is difficult to achieve because of the relatively small numbers of patients available to study prospectively. Despite these limitations, there are a number of studies that do provide evidence that the bone loss can be attenuated (if not stopped), and fractures decreased (if not prevented) through timely therapeutic intervention. In GIOP, it is necessary to distinguish between prophylactic measures in patients commencing corticosteroids (primary prevention) and treatment of established loss in patients on chronic, often low-dose corticosteroid therapy (secondary prevention) (Table 2).(23)

Table Table 2. Management of ARSBL
original image

It should be noted that therapeutic modalities for acute rapid bone loss have mostly been confined to GIOP because this was recognized as a disease entity for many years and only recently has there been an effort to examine therapeutic options in organ transplantation. Data for immobilization are sparse.

Calcium and vitamin D analogs

Multiple placebo-controlled studies(131,132) have demonstrated the importance of adequate calcium intake for both primary and secondary prevention of osteoporosis. Chronic calcium deficiency inevitably leads to skeletal demineralization and enhanced fracture risk.(131,132) The effects of calcium supplementation are maximized in patients in whom baseline intake is low.(131) The 1997 Consensus Development Conference on optional calcium intake recommended intakes of 1.5 g of “elemental” calcium per day for postmenopausal women,(133) although intake must be individualized. Risks of calcium supplementation are minimal, but those with a personal or family history of nephrolithiasis must be screened with 24-h urinary calcium determination.

Although calcium and vitamin D is routinely prescribed to all patients in the United States and worldwide after organ transplantation, except perhaps after renal transplant in patients with hypercalcemia, evidence for a benefit of this combination in reducing fracture is scarce. Measurement of serum 25OH-vitamin D is recommended, and values considered to be within normal limits (i.e., 15 ng/ml) may in fact represent subclinical vitamin D deficiency. Thus, one must ensure adequate intake, but hypercalcemia is to be avoided.

Calcitriol is effective in preventing bone loss after heart or lung transplantation but has to be continued long-term to maintain its effect.(134–137) The occurrence of hypercalcemia requires that serum calcium is routinely monitored. It is recommended that calcium and vitamin D therapy be used as adjunctive therapy with the more potent antiresorptive agents used to prevent high-turnover bone loss. Despite the widespread use of calcium and vitamin D and analogs as adjunct therapy with anti-osteoporotic drug administration, their use in post-transplant and GIOP is less efficacious than the N-containing bisphosphonates. In a recent study in GIOP, alendronate seemed to be superior to simple vitamin D and calcitriol in increasing spinal BMD and reducing vertebral fractures.(138) This effect probably applies to transplant bone disease as well.

Hormone replacement therapy

Estrogen has been shown to have some bone-sparing properties in GIOP,(139) and in one small randomized controlled trial (RCT) in patients with rheumatoid arthritis, a subgroup of patients showed modest increases in spine but not hip BMD(140); however, no fracture data are available. However, if a woman is amenorrheic and has no contraindications, hormone replacement therapy (HRT) can be administered for postmenopausal symptoms at least for the short term.(141) The recent Women's Health Initiative (WHI) report may make this option unfavorable unless the postmenopausal incidence of hot flashes is worrisome to the patient.(142,143) Androgen therapy for hypogonadal men is recommended provided it can be monitored and no contraindications exist. Data on its effect in these disorders are minimal and pertain to increases in BMD but not to fracture results.(144) Hypogonadism in males may be transient; therefore, androgen therapy should be periodically stopped and the clinical situation reassessed post-transplant. Topical testosterone is the preferred form of replacement, because circulating levels seem to be more physiological.(145,146)


There are a number of studies that have reported the use of calcitonin both in GIOP and in transplant patients, and the results have been mixed.(137,147–150) The common route of administration has been intranasal. Calcitonin is a powerful antiresorptive agent in vitro.


Most of the research with bisphosphonates, which applies to the disorders we are describing, has been conducted in GIOP. Adachi et al.(151) demonstrated that etidronate was effective in preventing bone loss in steroid-treated patients, but the fracture reduction was not statistically significant. The advent of the nitrogen-containing bisphosphonates provided an effective treatment for both prevention and treatment of GIOP. Both alendronate(25) and risedronate(152–155) have been shown, most convincingly, to cause increases in BMD and significantly reduce vertebral fracture in pooled data from prevention and treatment groups in premenopausal females, males, and postmenopausal females on glucocorticoids.

RCTs using risedronate studied patients initiating (the prevention study) or continuing (the treatment study) long-term glucocorticoid treatment. In the prevention study, the lumbar spine BMD of the risedronate 5 mg group increased significantly versus baseline and control by 3 months. In the treatment cohort, risedronate 5 mg increased BMD versus baseline and control at 6 and 12 months. In the prevention study, risedronate significantly increased BMD from baseline at the lumbar spine, femoral neck, and femoral trochanter in patients initiating glucocorticoids. In contrast, the placebo group (which was receiving calcium) lost BMD at both spine and hip.(154) In the treatment study, risedronate significantly increased BMD from baseline at the lumbar spine and femoral neck in patients taking long-term glucocorticoids.(155) Although the glucocorticoid-induced osteoporosis studies were not powered to detect vertebral fracture reduction, a significant 70% reduction (p = 0.01) in the incidence of new vertebral fractures was observed when the prevention and treatment data were combined for analysis.(152)

The initial alendronate study was different in that patients were stratified according to the time of initiation of therapy. Over 1 year, BMD increased significantly in the lumbar spine compared with the calcium and vitamin D group in the primary prevention group (less than 4 months of therapy). In the chronic users (longer than 12 months), there was an effect on increasing BMD from both alendronate and calcium and vitamin D therapy.(156) Post hoc analysis demonstrated that only postmenopausal women benefited in terms of spinal fracture reduction.(25) Intravenous formulations, such as pamidronate, have shown the most promise(157–163) because of the convenience of administration and ensuring compliance.

With the realization that a large proportion of the world's transplant patients are at various stages of bone loss, are continuing to fracture, and remain untreated, oral bisphosphonates are likely to be used increasingly. Unfortunately double-blinded placebo-controlled designs are lacking. Nevertheless, fracture reduction in liver and cardiac transplant patients has been impressive(153,157–163) and points the way for further and much needed large-scale controlled trials examining the efficacy of both oral and intravenous bisphosphonates in transplant patients.

New bone anabolic agents

A recent exciting newcomer to the therapeutic arena for osteoporosis is recombinant human parathyroid hormone (rhPTH). This drug represents an anabolic agent as opposed to a traditional antiresorptive drug, and its potential in these diseases should be large, especially where glucocorticoids are being or have been used. PTH, when administered intermittently and at low doses, is a potent stimulator of bone formation. Thus, it is most likely to reverse the bone loss and possibly restore the lost microarchitecture in GIOP. The same rationale applies to immobilization and stroke, which again have reduced bone formation. Indeed, Lane et al.(164) demonstrated impressive increases in BMD with rhPTH given to women on HRT and glucocorticoids. Prospective trials for fracture reduction are needed before final judgment can be made for the use of PTH in these disorders. Drawbacks include the daily subcutaneous injections required and the warning of the occurrence of osteosarcoma in rats, which occurred after excessive and life-term doses of PTH.

Future directions

The severe loss of bone and subsequent high risk of fracture in the above conditions justifies the need to prevent and treat this bone loss early and aggressively. The cornerstone of the prevention of osteoporosis in glucocorticoid-treated diseases will be to employ bone-sparing drugs. Likewise, in organ transplantation, administration of agents that are immunosuppressants to prevent organ rejection but are bone sparing will be needed. If their use is mandated, the lowest dose for the shortest possible time should be used. In immobilized patients with or without stroke, early mobilization and the development of “user friendly” hip protectors, which ensures improved compliance should be encouraged, to prevent fractures when these patients are mobilized. If the above cannot be achieved, it is urgent that RCTs be conducted with current therapies of antiresorbers and anabolic agents to provide robust evidence of their efficacy. It is also of high priority that physicians recognize and treat these conditions to reduce or prevent the devastating consequences of rapid bone loss and fracture. This education may be reinforced by development of suitable guidelines.


We have therefore described spectrum of diseases characterized by acute, rapid, and severe bone loss resulting in a high fracture incidence with devastating consequences. Ideally, patients need to be identified early, and preventative therapy initiated promptly to avoid the rapid bone loss and fractures. The most effective therapy to date, given the limitations of the studies, seems to be the bisphosphonates. However, forms of rapid bone loss that result from an osteoblastic defect and reduced bone formation may benefit from newer anabolic agents, such as rhPTH. Unless large scale clinical trials are initiated, which because of the relatively small numbers of patients transplanted need to involve multiple transplant centers, the answers to optimum clinical management will remain empiric. Effective guidelines need to be formulated and implemented so that the clinical manifestations and fractures prevented with effective therapies.


The authors acknowledge the generous support of the Bronx Veteran's Affairs Medical Center, Dept. of Medicine (Bronx). MZ acknowledges support of the National Institute of Aging (RO1-AG 14197–08) and the Department of Veterans Affairs (VA Merit Award). Fruitful discussions with Henry Bone are also acknowledged.