Osteoporosis occurs much less frequently in premenopausal women than in postmenopausal women. However, premenopausal osteoporosis can cause significant morbidity and premenopausal fractures are associated with increased risk of fractures later in life . It is thus important to recognize this entity and to conduct an evaluation for underlying causes of low-trauma fractures in premenopausal women. The diagnosis and management of osteoporosis in premenopausal women pose unique challenges and considerations compared to postmenopausal osteoporosis. In this chapter, we will review the epidemiology, diagnostic criteria, etiology, and management considerations for premenopausal osteoporosis. This topic has been reviewed in prior publications , and here we attempt to provide a summary of currently available data.
Overview of osteoporosis in premenopausal women
Estimates of the prevalence of premenopausal osteoporosis vary considerably, but it is clear that the prevalence of osteoporosis is substantially lower among premenopausal than postmenopausal women . Multiple studies demonstrate that fractures occur less frequently among premenopausal women . In one prospective study conducted in Dorset, England, the incidence of distal radius fracture was 10 per 10,000 population per year among premenopausal women, compared to a peak of 120 per 10,000 population per year in women older than 85 years . While rare, a history of premenopausal fractures appears to be a strong independent predictor of future fracture risk. In the Study of Osteoporotic Fractures, participants with premenopausal fracture(s) were almost 35% more likely to have a fracture after menopause than women without premenopausal fractures . The risk only changed slightly after adjusting for bone mineral density (BMD) and other confounders. A retrospective cross-sectional study of 1284 postmenopausal women in New Zealand reported an even greater risk of postmenopausal fractures after premenopausal fractures, with women who had self-reported fractures between ages 20 and 50 having a 74% increased risk of fracture after age 50 . In this study and others , premenopausal fractures remained a significant predictor of postmenopausal fracture risk after controlling for factors such as BMD, estrogen use, and maternal fracture history.
Definitions of osteoporosis in premenopausal women
Osteoporosis is a condition of impaired bone strength and increased fracture risk. Osteoporosis in premenopausal women can be most definitively diagnosed in the presence of low-trauma fracture(s). A low-trauma fracture is defined as that which occurs with force equivalent to a fall from a standing height or less. A stress or insufficiency fracture occurs with repetitive stress rather than a discrete traumatic event—and it is less clear which scenarios represent low-trauma events. A stress fracture or stress reaction in the context of normal stress suggests the presence of osteoporotic bone, while a stress fracture or reaction in the context of unusual stress may occur in normal bone. These definitions leave some room for interpretation by the patient and by the clinician and the “low-trauma” categorization may not be clear in all cases.
In some forms of primary osteoporosis, patients may have bone fragility related to bone quality defects, even in the absence of low BMD values. It is important to note that conditions such as osteomalacia, malignancy, and bone lesions can lead to pathologic fractures, and must be ruled out in cases of unusual fractures in order to diagnose osteoporosis.
Bone mineral density
BMD measurements in premenopausal women must be interpreted differently than those obtained in postmenopausal women. The International Society for Clinical Densitometry (ISCD) , International Osteoporosis Foundation (IOF) , and other experts recommend against diagnosing young women with osteoporosis based on BMD by dual-energy X-ray absorptiometry (DXA) alone, without a history of fragility fracture or secondary cause of osteoporosis. The condition in which patients have low BMD without fractures and without a clear underlying cause is termed idiopathic low BMD and will be discussed later in this chapter.
The ISCD BMD-based definition of premenopausal osteoporosis includes the presence of low BMD for age (Z≤−2.0) in addition to the presence of risk factors for fracture or secondary causes of osteoporosis . The ISCD recommends use of Z-scores (comparison to age-matched norms) instead of T-scores (comparison to young premenopausal norms) when interpreting BMD measures in premenopausal women, and thus works to avoid the diagnostic and treatment implications of reporting T-scores in these patients . A Z-score of ≤−2.0 should be interpreted as “below the expected range for age” and a Z-score of >−2.0 can be considered “within the expected range for age” . The category of “osteopenia” is not used in premenopausal women.
The IOF uses T-scores in its BMD-based definition of osteoporosis . The IOF definition of osteoporosis in young adults is a T-score of ≤−2.5 at the spine or hip in individuals who have completed growth and who have an ongoing secondary cause of bone loss or a chronic disorder known to affect bone mass. Of note, the ISCD also recommends use of T-scores rather than Z-scores in DXA interpretation for perimenopausal women .
Interpretation of bone mineral density in premenopausal women
The different emphasis on use of DXA measurements in osteoporosis definitions for premenopausal and postmenopausal women is based in part on a lack of information about how to interpret the risks conveyed by BMD measurements in premenopausal women. Postmenopausal osteoporosis can be diagnosed based upon low BMD by DXA even in the absence of fractures. Given the large quantity of observational and interventional data available regarding correlation of BMD by DXA with fractures in postmenopausal women, DXA measurements of BMD are important components of fracture risk prediction models that are used for therapeutic decision-making for postmenopausal osteoporosis. The same quantity and quality of data regarding BMD measurements by DXA and fracture risk are not available in premenopausal women. We cannot extrapolate recommendations regarding use of DXA measurements in a postmenopausal population to guide diagnosis and treatment decisions for premenopausal women .
Although fewer studies are available for premenopausal women, there may be a relationship between low BMD and fractures in this population. Several studies have shown that premenopausal women with Colles fractures had significantly lower BMD at the nonfractured radius , lumbar spine, and femoral neck , than premenopausal controls without fractures. Studies show mixed results for a relationship between BMD and stress fractures in premenopausal women. Stress fractures are a unique type of fracture related to repeated mechanical stress rather than acute injury. Some studies with female military recruits and athletes have demonstrated that stress fractures were associated with lower BMD compared to controls . However, in one study of US Military Academy cadets, the percentage of individuals with at least one stress fracture was higher in young women than men (19.1% vs 5.7%) and BMD was not associated with fracture risk in either population .
Given the lack of prospective studies correlating BMD to fracture incidence, as well as the low incidence of fractures in premenopausal women , the IOF and the ISCD recommend against the use of BMD by DXA as the only measure to guide diagnosis and treatment of osteoporosis in this population , and also recommend against use of DXA for screening for osteoporosis in premenopausal women . BMD measurements are indicated for premenopausal women with a history of low-trauma fracture and should be considered for women with conditions that result in increased bone loss and fracture risk, as will be discussed later in this chapter.
Idiopathic low bone mineral density
Idiopathic low BMD is a term that can be used to describe low BMD by DXA in premenopausal women in the absence of fractures or a clear underlying cause . As discussed, BMD findings alone do not define osteoporosis in these patients, and the associated fracture risk is unknown. DXA is an areal, two-dimensional measurement of BMD and is affected by bone size, and it can thus be smaller in individuals with smaller bones . This raises the question of whether healthy, young women without other medical conditions or fractures, who have an isolated finding of low BMD by DXA, truly have low volumetric BMD, reduced bone strength, and ultimately increased fracture risk .
Several studies suggest that these patients may have abnormal bone microarchitecture . More specific evaluation of volumetric bone BMD and bone microarchitecture can be achieved via advanced imaging techniques such as central and peripheral computed tomography (CT) and high-resolution imaging of bone biopsy samples. One study suggests that healthy, normally menstruating premenopausal women with idiopathic low BMD (when compared with healthy controls with normal BMD) have thinner, more widely spaced and heterogeneously distributed trabeculae and thinner cortices as well as decreased estimated bone strength . These findings in women with idiopathic low BMD were comparable to those of a similar group of premenopausal women with fragility fractures, and the similarities remained even after correcting for the smaller bone size in the women with idiopathic low BMD and no fractures. Low areal BMD by DXA was noted to be a reasonable predictor of low volumetric BMD by central quantitative CT . Limitations of this study include small sample size and possible ascertainment bias. Many in the idiopathic low BMD group had a family history of osteoporosis (84%), childhood fractures (26%), or high-trauma adult fractures (16%), which may have affected their study enrollment .
In another observational study, constitutionally thin premenopausal women [body mass index (BMI)<16.5 kg/m 2 , normal menstruation, and no evidence of anorexia nervosa or a secondary, systemic disorder] were found to have smaller bone size, lower lumbar and femoral BMD, and diminished breaking strength compared to age-matched, normal-weight controls .
These studies are small, observational, and may not be generalizable to all premenopausal women, but they do suggest that asymptomatic, isolated low BMD in premenopausal women is associated with structural abnormalities that may be a risk factor for development of symptomatic osteoporosis. Again, however, there is insufficient evidence to use low BMD, alone, to guide therapeutic decisions in this population, as is discussed later in this chapter.
Special considerations in interpreting bone mineral density results in premenopausal women
Peak bone mass accrual
When interpreting BMD in premenopausal women, one must consider whether peak bone mass has been achieved. Bone mass continues to increase from childhood to young adulthood, and determinants of age at peak bone mass include ethnicity , gender , body size, age of menarche , and skeletal site. In girls, bone mass accrues most rapidly from age 11 to 14 years . By 2 years after menarche, rate of accrual decreases and at least 90% of peak bone mass is achieved by late teen years , with the possibility of some additional bone accrual from age 20 to 29 years . Age of peak bone mass in women has been suggested to be site-specific , occurring in the femur during the third decade of life and at the forearm and spine at about age 30 .
Genetic predisposition or processes that interfere with bone mass accrual (e.g., medications, illness) can lead to a below-average peak bone mass (and BMD by DXA). Both scenarios may result in low premenopausal BMD as well as less “bone reserve” in the estrogen-deficient years after menopause. Some conditions may be temporary and thus not identifiable at the time of premenopausal (or postmenopausal) osteoporosis evaluation. These factors have an uncertain effect on bone microarchitecture and strength. In addition, there are no published studies evaluating bone quality in women with low peak bone mass due to secondary causes of bone loss compared to bone quality in women with a genetic predisposition to low peak bone mass (primary forms of osteoporosis).
Physiologic changes associated with pregnancy and lactation
Pregnancy and lactation are states of high calcium demand and are associated with substantial changes in calcium and bone metabolism. Large physiologic changes in bone mass are expected, including rapid asymptomatic decrease in spine and hip BMD during normal pregnancy and lactation (if the woman chooses to breastfeed) and recovery during and after weaning. These expected changes must be considered when interpreting BMD measurements obtained around this time.
Calcium demands from the growing fetus cause a decline in bone mass by 3%–5% over a typical pregnancy , as shown by studies examining BMD in women before, during, and after pregnancy using techniques such as DXA as well as distal radius peripheral quantitative computed tomography and heel ultrasound . This decline occurs despite a significant increase in the production of active vitamin D (1,25(OH) 2 ) and subsequent doubling of intestinal calcium absorption .
During lactation, calcium metabolism changes substantially. Intestinal hyperabsorption of calcium stops, and breast milk production is associated with an increase in calcium demand to 300–400 mg/day. This demand is met by PTHrP secretion from the mammary glands, which causes calcium mobilization from the skeleton . Large bone mass declines during lactation have been consistently documented in several longitudinal studies. Over the first 6 months of lactation, BMD declines by 3%–10% at the lumbar spine and 2%–4% at the hip . Longer periods of lactation as well as postpartum amenorrhea are associated with greater bone loss . The duration of lactation is also associated with changes in plasma bone turnover marker levels . As a frame of reference, there is an average bone loss of 1%–3% per month during lactation compared to 1%–3% per year immediately following menopause . However, fractures during lactation are extremely rare.
Postpartum bone recovery
In nonlactating women, bone mass remains stable or increases postpartum and increases may be aided by calcium supplementation . In lactating and nonlactating women, bone mass recovery appears to be related to duration of lactation and return of menses . In lactating women, even with ongoing lactation, recovery often occurs after the 6-month postpartum period . At this point, the infant often starts to eat solid food, with decreased requirements for milk production and thus decreased calcium mobilization requirements. In addition, on average, menses resume at 6- to 8-month postpartum .
Bone mass recovery after weaning appears to be site-specific, as has been documented in both human and rodent studies that show complete recovery at the spine, but only partial or delayed recovery at the femur . The human studies include a limited (12–20 month) postpartum follow-up; longer term studies would be needed to determine whether full recovery is also achieved at nonvertebral sites.
Potential long-term structural changes
Human and animal studies suggest that there may be an increase in bone size associated with lactation . In one rodent study, multiple reproductive cycles were associated with both trabecular degeneration and structurally positive changes in cortical bone, including increased cortical area . These beneficial structural changes may confer a protective effect against future estrogen deficiency–related bone loss .
Effect of parity and lactation on postmenopausal bone mineral density and fracture risk
While some studies suggest incomplete bone mass recovery at some skeletal sites in the postpartum/postweaning period, most epidemiological studies suggest that there is no net negative effect of changes during and after pregnancy and lactation on postmenopausal bone mass or long-term fracture risk.
In some studies, pregnancy and lactation history have been associated with decreased risk of osteoporosis or fracture . In addition, grand multiparity, repeated cycles of pregnancy and lactation without recovery intervals, and extended periods of lactation have not been associated with lower BMD by DXA in cross-sectional comparison to nulliparous controls .
In one analysis of data from the Women’s Health Initiative study, 92,980 multiracial women in the United States, for whom pregnancy and lactation history were available, had incident fracture rate assessment over a mean of 7.9 years . After adjusting for factors such as years since menopause, family history of fracture, BMI, estrogen use, and calcium and vitamin D supplementation, there was no significant association between incident hip fracture and number of pregnancies, maternal age at first birth or at first lactation, number of children breastfed, or total duration of breastfeeding. Compared to women who had never breastfed, women who breastfed for at least 1 month had a 16% lower risk (HR 0.84; 95%CI: 0.73–0.98). Another systemic review also found a positive effect of parity on hip BMD, compared to nulliparity .
Regional differences may exist, and some studies have documented contrasting effects of pregnancy and lactation on bone mass . Studies from Turkey and China found that duration of lactation was negatively associated with postmenopausal BMD . A study of the mestizo population in Mexico found that 24 months or more of breastfeeding was associated with postmenopausal osteopenia and osteoporosis . Differences in nutritional intake and socioeconomic factors may play a role in the differences noted among studies.
Pregnancy- and lactation-associated osteoporosis
Pregnancy- and lactation-associated osteoporosis (PLO or PLAO) is a very rare, severe early presentation of osteoporosis occurring during late pregnancy or during lactation. Women with PLO sustain low-trauma or spontaneous fractures, most commonly vertebral fractures that lead to symptoms of significant back pain and height loss . Most cases of PLO occur in primigravid women who are otherwise healthy without a known predisposing condition . Prepregnancy BMD is, thus, generally unknown for these patients, but BMD by DXA at the time of presentation is usually extremely low with Z-scores often reported at less than −3.0.
The majority of women with PLO present with vertebral fractures (89%–93%) , but other types of fractures have also been reported. Presentation with femoral neck fracture(s) during or after pregnancy has been described in a minority of cases and is sometimes described using the term “transient osteoporosis of the hip.” It is not clear whether PLO-associated fractures of the hip occur in the setting of an additional insult (such as infection, inflammation, or ischemia) that predisposes to site-specific marrow edema and bone weakness , or whether this condition simply represents an alternative location for the trabecular bone weakness that defines PLO. Unilateral and bilateral progressive avascular necrosis of the femoral head, with or without fractures, has also been associated with pregnancy , but it remains unknown if transient osteoporosis of the hip is a precursor to avascular necrosis of the femoral head, or whether avascular necrosis in pregnancy is a separate clinical entity.
The etiology or etiologies of PLO are unknown. In some cases, women may enter pregnancy with a diagnosed or undiagnosed skeletal deficit related to a primary/genetic etiology, or a secondary etiology such as a medical condition or medication exposure. Preexisting skeletal deficits in these contexts may then be exacerbated by physiologic bone changes of pregnancy and lactation, leading to excessive bone fragility and fractures.
Some reports describe women who have PLO in the setting of preexisting conditions or medication exposures that might affect their BMD. Documented cases include patients with a history of exposure to low-molecular-weight heparin or heparin , glucocorticoid use , antiepileptic medication exposure , vitamin D deficiency , thyroid hormone treatment at suppressive doses , prolactinoma , anorexia nervosa , and celiac disease .
In support of a primary or genetic bone-specific etiology in some, many women whose cases are discussed in case reports and cohorts of PLO are otherwise healthy, with no known predisposing condition and with normal biochemistries. Family history of osteoporosis is common and specific genetic etiologies such as osteogenesis imperfecta (OI) and LRP5 mutations have been reported . Additionally, at the tissue level, women with PLO have been shown to have a particularly low bone remodeling state, suggesting the possibility of an underlying osteoblast functional deficit . In this study examining microarchitectural changes in premenopausal women with osteoporosis, transiliac crest bone biopsy samples were obtained greater than 1-year postpartum from 15 women with PLO (80% with vertebral fractures), 63 premenopausal women with unexplained fractures (IOP), and 40 healthy controls. The PLO and IOP groups were similar with regard to age, BMI, body fat percentage, age of menarche, parity, and age at first pregnancy. PLO subjects had lower spine BMD than IOP women and controls and tended to have lower cortical width on biopsies and lower bone turnover. All of this suggests an underlying bone formation defect in PLO, with significantly lower bone formation rate and tissue-level mineral apposition, in addition to lower bone turnover markers (PINP and CTX) .
Given that PLO is rare, we have limited information about its natural history. Case reports and limited case series have reported on follow-up in untreated women for several years after the incident fracture . Some suggest BMD recovery postpartum, similar to recovery seen in healthy women . However, these patients appear to have a high subsequent fracture risk. In one single-center prospective study of 107 PLO women, 24% had subsequent fractures during a median 6-year follow-up period . There was a modest correlation between number of initial fractures and future fracture risk ( r =0.56, P =.003). A total of 20% of the 30 women who had a subsequent pregnancy reported fractures associated with their next pregnancy, even though 76% of these patients had previously received bisphosphonates or teriparatide treatment . Many case reports document no recurrent fractures in subsequent pregnancies but repeated vertebral fracture has also been documented .
Evaluation of premenopausal women with low-trauma fracture and/or low bone mineral density
Primary osteoporosis diagnosed in adulthood
Conditions that are associated with abnormal skeletal development and bone fragility in childhood can present heterogeneously and with widely varying degrees of severity. In rare cases, these conditions may lead to symptoms and/or diagnosis in early adulthood, rather than in childhood. Examples of such causes of primary osteoporosis include OI , hypophosphatasia (associated with osteomalacia) , osteoporosis-pseudoglioma syndrome or LRP5 mutations , Marfan and Ehlers-Danlos syndromes. Severity of disease, age at presentation, and associated clinical features may prompt an evaluation for a primary cause of osteoporosis.
Secondary causes of premenopausal osteoporosis
When premenopausal women present with low-trauma fractures, an underlying cause of bone weakness can usually be identified. In one study of osteoporosis in young adults in Minnesota, 90% of individuals had a secondary cause of their osteoporosis . Other studies from tertiary care referral centers have identified that 48%–53% of cases of premenopausal osteoporosis had secondary causes . It is possible that more unexplained cases are referred to tertiary care referral centers, leading to a higher proportion of patients with unidentified secondary causes in this group.
Underlying conditions may interfere with peak bone mass accrual or may cause bone losses after peak bone mass has been achieved. Secondary causes of osteoporosis include medication exposure (particularly glucocorticoids), various conditions causing estrogen deficiency, gastrointestinal disorders causing malabsorption, and thyrotoxicosis. Some conditions such as anorexia nervosa or inflammatory bowel disease may lead to increased bone fragility through multiple mechanisms (i.e., malnutrition, estrogen deficiency, other hormone abnormalities, inflammation). Table 38.1 organizes some of these secondary causes of osteoporosis.
|Any disease affecting bone mass accrual during puberty and adolescence|
|Primary osteoporosis/connective tissue diseases|
|Relative energy deficiency in sports (previously called female athlete triad)|
|Pathologic hyperprolactinemia from a nonhypothalamic-pituitary cause|
|Anorexia nervosa (multifactorial cause of osteoporosis including several nutritional and hormonal abnormalities)|
|Other endocrinopathies and abnormalities of calcium metabolism|
|Cortisol excess/Cushing syndrome|
|Postoperative states (e.g., Roux-en-Y gastric bypass)|
|Inflammatory bowel disease|
|Significant vitamin D, calcium, and/or other nutrient deficiency|
|Systemic lupus erythematosus|
|Medications (not all have been studied in premenopausal populations)|
|Medication-related amenorrhea and suppression of ovulation|
|GnRH agonists (when used to suppress ovulation)|
|Depot medroxyprogesterone acetate|
|Chemotherapy leading to amenorrhea|
|Calcineurin inhibitors (e.g., cyclosporine)|
|Antiepileptic drugs (particularly cytochrome P450 inducers such as carbamazepine and phenytoin)|
|Chemotherapeutic drugs with bone-related toxicity (particularly high-dose methotrexate)|
|Heparin (Unfractionated heparin is associated with both BMD loss and increased fracture risk . Low-molecular-weight heparin exposure for >3 months is associated with decrease in BMD, but no known increase in fracture risk .)|
|Proton pump inhibitors|
|Vitamin A excess|
|Diabetes (Type 1 and 2)|
|HIV infection and/or medications|
|Alcohol use disorder|
|Liver disease (particularly cholestatic liver disease)|
Premature ovarian failure, as from autoimmune diseases, chromosomal abnormalities such as fragile X syndrome and Turner syndrome, and toxins such as chemotherapy and radiation, can cause osteoporosis in young women as well. These etiologies are not discussed in this chapter focusing on premenopausal osteoporosis.
It is important to conduct a comprehensive evaluation for secondary causes of premenopausal osteoporosis, since specific treatment of several of these conditions has been associated with BMD gains. Data for premenopausal women are not always available, but examples of conditions in which treatment of the disease may be beneficial for bone health include hypercortisolism (endogenous and iatrogenic), hyperparathyroidism , celiac disease , estrogen deficiency, hypercalciuria , and Crohn disease .
Approach to evaluation of premenopausal osteoporosis and/or isolated low bone mineral density
Detailed clinical history and thorough physical examination are important first steps in the evaluation of premenopausal women with low-trauma fractures and/or low BMD, since these may help to identify an underlying primary or secondary cause of bone fragility. History can include questions about current and prior illnesses, medication use, diet and exercise, gastrointestinal symptoms, nephrolithiasis, and family history of osteoporosis and fractures. It is also important to obtain a complete menstrual and reproductive history, including history and timing of lactation, if applicable. Physical examination may help to evaluate for certain secondary causes of osteoporosis including hypercortisolism, hyperthyroidism, and eating disorders. Features such as joint hyperlaxity, blue sclerae, and dentinogenesis imperfecta might suggest an underlying primary cause of osteoporosis or a connective tissue disorder. Additionally, laboratory evaluation should be conducted, including an electrolyte panel, complete blood count, 25-OH vitamin D (if severely deficient, investigation for osteomalacia may be pursued), parathyroid hormone (PTH), liver function tests including alkaline phosphatase, thyroid stimulating hormone (TSH), and 24-hour urine calcium and creatinine. Cortisol assessment may be considered, if clinically indicated. Abnormalities on initial evaluation can guide additional testing to identify secondary causes (see Table 38.2 ).
|Initial laboratory evaluation|
|Additional laboratory tests based on clinical scenario|
Although screening BMD is not recommended in premenopausal women, if a patient presents with a finding of unexplained very low BMD, this should lead to a similar clinical evaluation for secondary causes. In those with fractures or isolated low BMD, serial BMD measurements can help to determine whether there is ongoing bone loss. Continued BMD decline indicates the need for a continued search for underlying causes that may affect clinical management.
Bone turnover markers
Elevations of bone turnover markers such as C-telopeptide, N-telopeptide, bone-specific alkaline phosphatase, and osteocalcin may be caused by multiple processes; thus, these markers have limited utility in the evaluation of premenopausal osteoporosis. Elevations may indicate active bone modeling in young adulthood, ongoing bone loss, variations across the menstrual cycle , or bone remodeling after a fracture. The 2020 American Association of Clinical Endocrinologists/American College of Endocrinology clinical practice guidelines recommend considering evaluation of bone turnover markers to evaluate response to antiresorptives in premenopausal women, but do not make recommendations for use of these markers as part of an initial evaluation .
Transiliac crest bone biopsies may rarely be useful in elucidating the mechanism of low BMD or fragility fractures. Analysis of bone biopsy samples allows examination of detailed microarchitectural features of bone, and tetracycline labeling prior to the biopsy procedure enables evaluation of dynamic parameters including bone formation rate. Bone biopsy may be helpful in differentiating between different types of renal osteodystrophy, ruling out osteomalacia, and uncovering rare causes of bone fragility associated with marrow changes, such as Gaucher disease or mastocytosis.
Idiopathic osteoporosis is a term used to describe the condition of low-trauma fractures in premenopausal women who have no known etiology of bone fragility found after a thorough clinical and laboratory investigation. Based on prior reports, the mean age at IOP diagnosis is 35 years , and IOP can manifest as a single low-trauma fracture of the hip, spine, or long bone or as multiple fractures (vertebral and nonvertebral) occurring over 5–15 years . Women with IOP are predominantly Caucasian and have a family history of osteoporosis . They also may have lower weight and height than controls .
On a structural level, the bones of women with premenopausal IOP have microarchitectural deficiencies compared to controls. As assessed by both high-resolution peripheral quantitative computed tomography and bone biopsy , these individuals have lower volumetric BMD, fewer and more heterogeneously distributed trabeculae, greater trabecular separation, thinner cortices, and reduced bone stiffness assessed by finite element analysis. These differences persist after controlling for bone size, age, and BMI .
Women with IOP, by definition, have normal gonadal function and menstruation. Studies have tested hypotheses related to more subtle estrogen exposure differences. Lower follicular phase estradiol levels in IOP women were noted in one study , but not in others . Thus there has been no consistent documentation of estrogen deficiency as a potential etiology of IOP.
Tissue-level bone remodeling rates are quite variable in IOP women, suggesting a heterogeneity in the pathogenesis of fractures in these patients . In a subgroup of IOP women with low bone formation rate and the most severe microarchitectural deficits, IGF-1 (insulin-like growth factor) levels were elevated, suggesting the possibility of osteoblast resistance to IGF-1. Women with frank hypercalciuria were excluded from the study for idiopathic osteoporosis, but a high bone remodeling group had a pattern resembling idiopathic hypercalciuria, with high serum 1,25(OH) 2 D and mildly elevated 24-hour urinary calcium .
Treatment approach for premenopausal women with low-trauma fractures or low bone mineral density
There are limited high-quality data available to guide management decisions for women with premenopausal osteoporosis and low BMD. A general approach to management incorporates lifestyle modifications to optimize nutrition and bone health and also addresses underlying conditions and medication exposures that may contribute to bone fragility, whenever possible. Pharmacologic treatment is only considered in certain cases, particularly for women with a history of significant fragility fractures and in some cases for women with very low bone mass in the context of an ongoing underlying cause that cannot be adequately addressed.
We can consider management options in three groups of premenopausal patients: women with idiopathic low BMD and no low-trauma fractures, women with low-trauma fracture(s) without an identifiable underlying cause, and women with low BMD or low-trauma fracture(s) with a known secondary cause.
In young women with isolated idiopathic low BMD (low BMD, no prior low-trauma fractures, and no clear underlying cause), the diagnosis of “osteoporosis” does not apply, and pharmacologic therapy is rarely indicated. Based on key principles regarding interpretation of bone mass measurements in premenopausal women (discussed earlier), low BMD alone should not be used to diagnose osteoporosis or to make therapeutic decisions in premenopausal women. This is because of the absence of longitudinal data to correlate BMD with short-term fracture risk in this population, along with the overall low fracture rates in premenopausal women. Although bone microarchitectural abnormalities have been documented in women with isolated idiopathic low BMD , current evidence supports an expectation for stable BMD over time , and a low short-term risk of fracture. Ongoing monitoring with sequential DXA scans every 1–2 years may provide guidance as to the need for continued evaluation for secondary causes and/or the need for pharmacologic therapy if ongoing BMD declines are noted, particularly in the context of extremely low BMD scores (e.g., Z-score<−3).
In premenopausal women with idiopathic osteoporosis characterized by history of low-trauma fractures and no identifiable secondary cause of bone fragility, pharmacologic treatment is more likely to be considered on a case-by-case basis and may be of benefit. Because data are limited regarding risks and benefits of osteoporosis medications for these patients, it is our opinion that osteoporosis medications should be reserved for women with a history of major fractures of the spine or hip, multiple fractures, and/or declining BMD.
Finally, if a secondary cause of impaired bone strength is known, the first step in management should be to address this underlying cause, if possible. Examples of interventions expected to improve skeletal health include minimization or cessation of medications that might affect bone mass, treatment of hyperthyroidism, parathyroidectomy for hyperparathyroidism , management of hypercortisolism related to Cushing syndrome , dietary modification for patients with celiac disease , and nutritional rehabilitation and promotion of weight gain for patients with anorexia nervosa . Thiazides may provide a skeletal benefit when used to address idiopathic hypercalciuria , although limited data are available specifically in relation to premenopausal women. If the underlying cause can be addressed or eliminated, patients may have significant improvements in bone strength, preventing the need for pharmacologic treatment. However, there are cases in which the underlying cause cannot be eliminated and leads to an ongoing risk of bone loss and/or fractures. Pharmacologic treatment may be considered for such patients.
Lifestyle modifications and dietary supplementation
Several nonpharmacologic interventions may be recommended for patients with premenopausal fractures and low bone mass. Smoking cessation and avoidance of excessive alcohol intake are commonly recommended for all patients, despite lack of data to confirm benefits of these interventions for BMD or fracture risk . Randomized trials evaluating physical activity in premenopausal osteoporosis do not include fracture endpoints, but physical activity, including regimens with resistance training, may prevent BMD losses in this population, and can be recommended as well . In a recent randomized trial in premenopausal women with breast cancer who were at risk for treatment-related bone loss, who completed a 12-month exercise intervention with resistance training and aerobic exercise, and who maintained lean muscle mass, had less BMD loss at the lumbar spine . Exercise regimens can be recommended on a case-by-case basis, with the caution that excessive exercise can also be detrimental to bone strength, via weight loss and/or hypothalamic amenorrhea.
Calcium and vitamin D intake are also recommended, although the optimal amount for premenopausal women is uncertain. The 2011 Institute of Medicine guidelines suggest a daily intake of 1000 mg of dietary and supplemental calcium and 600 IU of vitamin D in this population . More recently, the US Preventive Services Task Force (USPSTF) has concluded that there is insufficient evidence to assess the risks and benefits of supplementation for primary prevention of fractures in premenopausal women . Ultimately, recommendations for optimal calcium and vitamin D intake can be based on individual clinical scenarios, taking into account fracture history, measures of calcium metabolism, and vitamin D levels.
Combination oral contraceptives
Estrogen therapy appears to have beneficial effects on bone mass in premenopausal women with estrogen deficiency , and can be recommended in these patients. On the other hand, there is no clear evidence of benefit of oral estrogen/progestin therapy for patients with anorexia nervosa, which has multifactorial effects on bone mass . Skeletal complications of anorexia nervosa, a complex condition, are discussed in Chapter 51, Skeletal health after bariatric surgery. In a case-control study from the United Kingdom that included 12,970 premenopausal women, women without fractures were significantly more likely to have had a history of oral contraceptive use . There is inconclusive evidence, however, that oral contraceptives decrease fracture risk . In healthy premenopausal women without known estrogen deficiency, most studies have shown that use of oral contraceptives has no effect on bone mass or fracture risk . One study has documented lower BMD at the spine and whole body in young adult women using oral contraceptives at any dose for >24 months, compared with nonusers , while another study suggests that oral contraceptive use (20 µg estradiol dose) may impair peak bone mass accrual in women aged 19–23 years .
Selective estrogen receptor modulators
Raloxifene, a selective estrogen receptor modulator, is FDA approved for the treatment of postmenopausal osteoporosis and is associated with decreased vertebral fracture risk and improvements in BMD at the spine and femoral neck in postmenopausal women . In contrast, selective estrogen receptor modulators should not be used to treat bone loss in premenopausal women, since they can block the effect of estrogen on bone and exacerbate bone loss .
Oral and intravenous bisphosphonates are associated with BMD improvements in premenopausal women with several different conditions associated with bone loss including breast cancer treatment, anorexia nervosa, and glucocorticoid use . Studies have also shown BMD improvements with bisphosphonate use in young adults with conditions such as cystic fibrosis, OI, and thalassemia, but these studies have not specifically examined premenopausal women . Two bisphosphonates, alendronate and risedronate, are approved by the United States Food and Drug Administration (US FDA) for use in premenopausal women receiving glucocorticoids, given favorable short-term BMD outcomes. Further discussion is provided in Section 22.214.171.124 .
While bisphosphonate use may lead to BMD improvements for premenopausal women in specific clinical scenarios, fracture data are rarely available. Complications such as osteonecrosis of the jaw and atypical subtrochanteric femoral fractures are potential risks of long-term use of these agents . When bisphosphonates are initiated at a young age, patients may have more prolonged medication exposure. Thus the process of therapy initiation should include careful consideration of the potential risks of long-term use, as well as discussion of plans for duration of bisphosphonate treatment with the aim of treating for the shortest length of time possible.
Bisphosphonates carry a Category C rating for safety in pregnancy from the US FDA. They have been shown to cause toxic effects in pregnant rats, accumulate in the skeleton, cross the placenta, and accumulate in the fetal skeleton . Data on risk in humans are inconsistent. One report in humans documented transient neonatal hypocalcemia and bilateral talipes equinovarus in two neonates , but several other reports have documented no adverse maternal and fetal outcomes . The label for alendronate recommends that it “should be used during pregnancy only if the potential benefit justifies the potential risk to the mother and fetus” . In general, it is recommended that contraception be utilized during bisphosphonate use, and bisphosphonates should be used cautiously for women who may have future pregnancies, as these medications can remain in the skeleton for years and can have potential adverse effects even after discontinuation.
Denosumab is a RANK ligand inhibitor that is FDA approved for the treatment of osteoporosis in men and postmenopausal women. The safety and efficacy of this medication have not been defined in premenopausal women, and the current label states: “Pregnant women and females of reproductive potential: Prolia may cause fetal harm when administered to pregnant women. Advise females of reproductive potential to use effective contraception during therapy, and for at least 5 months after the last dose of Prolia” .
Teriparatide or PTH(1–34), a recombinant form of PTH, has been FDA approved for the treatment of osteoporosis in postmenopausal women and men who are at high risk for fracture, as well as for patients requiring sustained systemic glucocorticoid therapy who have osteoporosis or are at high risk for fracture. The current label for teriparatide does not specifically recommend against use in women of reproductive potential, but states: “There are no available data on FORTEO use in pregnant women to evaluate for drug-associated risk of major birth defects, miscarriage, or adverse maternal or fetal outcomes” .
BMD improvements have been observed with use of teriparatide in premenopausal women with idiopathic osteoporosis as well as a number of specific secondary conditions associated with bone loss . As is the case with bisphosphonates, there are no available data to demonstrate that teriparatide reduces fracture risk in these populations.
Premenopausal women treated with teriparatide for glucocorticoid-induced osteoporosis (GIOP) demonstrated greater improvements in lumbar spine BMD compared to those treated with alendronate (see Section 126.96.36.199 ) . In a small, placebo-controlled study of 21 women with anorexia nervosa and T-score≤−2.5 (mean age 47 years), women treated with teriparatide had significant improvements in spine BMD at 6 months, compared to women who received placebo (spine BMD increase 6.0% vs 0.2%; P <.01) . In one study looking at patients with ovarian suppression at risk for bone loss (premenopausal women treated with nafarelin, a GnRH analogue, for endometriosis), PTH(1–34) at doses of 40 μg/day (higher than the current recommended dose of 20 μg/day) prevented bone loss, with 2.1% improvements at the spine, compared to a 4.9% BMD decline in women not treated with PTH(1–34) ( P <.001) . It is not clear how these results would apply to women with baseline osteoporosis and women with osteoporosis and normal gonadal status.
In a small observational study of premenopausal women with IOP, patients treated with teriparatide 20 µg daily had average BMD increase of 10.8% at the lumbar spine and 6.2% at the total hip (both P <.001) after 18–24 months of treatment . However, four women had little or no improvement in BMD and were considered nonresponders. This subset of nonresponders had markedly lower tissue level baseline bone formation rate. Additionally, analyses of serum bone turnover markers over the course of treatment suggested that nonresponders lacked a clear “anabolic window” (an increase in bone formation that precedes the increase in bone resorption), which is usually seen in response to teriparatide . C-telopeptide, a resorption marker, rose comparably in responders and nonresponders, while the peak rise in P1NP (N-terminal propeptide of procollagen type 1), a bone formation marker, was blunted and delayed in the nonresponders. In a randomized double-blind placebo-controlled trial of teriparatide 20 µg daily in 41 premenopausal women with IOP, subjects who were randomized to receive teriparatide for the first 6 months of the study gained significantly more BMD at the spine than those receiving placebo (5.5%±4.2% vs 1.8%±4.0%; P =.01) . Similar to the prior observational study, patients followed an open-label extension as part of this study had variable response and some women were nonresponders .
Teriparatide product labeling specifies that use of drug for more than 2 years during a patient’s lifetime is not recommended . Because the long-term effects of teriparatide in young women are not known, teriparatide should only be considered in premenopausal patients with the highest fracture risk or ongoing recurrent fractures. Continued bone growth is a contraindication to teriparatide use, thus documentation of fused epiphyses is recommended in patients less than 25 years of age prior to consideration of teriparatide treatment.
In postmenopausal women, BMD declines after cessation of teriparatide in patients who are not treated with subsequent antiresorptive medications . Hormone replacement therapy may prevent bone loss after completion of teriparatide treatment in postmenopausal women , but women with IOP do not appear to be protected from bone loss after teriparatide cessation by their normal gonadal status. In one study, BMD declined by 4.8%±4.3% ( P <.001) in premenopausal women with IOP who did not receive a follow-up antiresorptive after teriparatide . This suggests that antiresorptive therapy is needed to prevent bone losses after teriparatide cessation, even in women without identified secondary causes of ongoing bone loss.
Abaloparatide is a PTHrP analogue that is FDA approved for the treatment of osteoporosis in postmenopausal women with high fracture risk. There are no data on the safety or efficacy of abaloparatide in premenopausal women, and the current label states that “TYMLOS is not indicated for use in females of reproductive potential” .
Romosozumab, FDA approved in 2019 for treatment of postmenopausal women at high risk of fractures, is a monoclonal antibody that inhibits the action of sclerostin. There are currently no data on safety and efficacy of romosozumab for premenopausal patients, and the current label states that romosozumab is “not indicated for use in women of reproductive potential” .
Pharmacologic therapy in specific clinical scenarios
Pregnancy- and lactation-associated osteoporosis
PLO is a very severe presentation of early-onset osteoporosis. Because it is quite rare, guidance regarding management is based on a small number of case reports and observational studies.
Therapeutic approaches for PLO are often implemented during the very dynamic period of bone recovery from pregnancy- and lactation-associated BMD losses (see Section 188.8.131.52 ). In healthy women, BMD gains, even up to 10%–20%, are expected as part of the natural recovery that occurs postpartum and postweaning . Thus, without placebo-controlled trials, it is difficult to assess the effects of therapeutic interventions for PLO in comparison to the expected natural course of BMD recovery postpartum and postweaning.
Several case reports have documented that patients with PLO can have improvement in BMD with bisphosphonate use, suggesting a possible role for bisphosphonates after delivery and weaning . In an uncontrolled study of five women with PLO-associated vertebral fractures who were treated with bisphosphonates within 2 years of presentation, a 23% improvement in spine BMD was observed at 24 months . In a retrospective study from France, among 18 patients with PLO who had monitoring of BMD during a mean 2.5-year follow-up, 7 patients received bisphosphonates and had an annual mean BMD improvement of 10.2% at the spine and 2.6% at the femoral neck . In comparison, 4 patients who received teriparatide had an annual improvement of 14.9% at the spine and 5.6% at the femoral neck, and 5 patients who received no treatment had an annual mean 6.6% BMD gain at the spine and 2.3% BMD gain at the femoral neck. These results suggest that both bisphosphonates and teriparatide may augment natural BMD recovery in women with PLO.
Treatment with teriparatide appears to lead to BMD improvements in patients with PLO in several case reports . In one observational study, BMD changes were compared between 27 women with PLO who were treated with teriparatide and 5 women with PLO who chose not to receive treatment . Patients treated with teriparatide had greater BMD improvements at the lumbar spine at 12 months (16%±7%) than patients who did not receive teriparatide (8%±7%). These results again provide evidence that teriparatide can augment natural BMD recovery in women with PLO. It is unclear whether patients treated with teriparatide for PLO require antiresorptive treatment to prevent bone loss after cessation of teriparatide treatment.
Ultimately, the role of bisphosphonates and teriparatide remains unclear in the absence of randomized controlled trials and long-term follow-up of effects on fracture risk.
Osteogenesis imperfecta in adults
While many studies have addressed therapeutic approaches for OI in childhood, a more limited number of studies investigate treatment approaches in adults with this genetic condition . Studies are generally small, do not definitively address fracture endpoints, and do not specifically evaluate a female population.
Bisphosphonate therapy has been evaluated in adult patients with OI in both randomized and nonrandomized studies . Both oral and IV bisphosphonates are associated with BMD improvements at the spine and hip. In one observational nonrandomized study, an intravenous bisphosphonate (pamidronate) appeared to be associated with a decreased ratio of pretreatment to posttreatment fractures in adults with types iii/iv OI compared to no treatment ( P =.05), but this ratio was not lower with use of oral bisphosphonates in patients with types iii/iv OI, nor with use of oral or intravenous bisphosphonates in patients with type i OI .
There have been a few randomized controlled trials examining teriparatide treatment in OI patients. In one study of 79 adult (predominantly type i) OI patients, significant increase in BMD at the spine and hip was seen in patients treated with teriparatide over 18 months, but no difference was seen in self-reported fractures . In another multicenter randomized trial comparing treatment with teriparatide to neridronate, an amino-bisphosphonate, in adult OI patients, teriparatide treatment was associated with larger BMD gains and a trend for reduced fractures during follow-up ( P =.1) .
Hypophosphatasia in adults
Hypophosphatasia is an inborn error of metabolism due to mutation of the gene encoding tissue nonspecific alkaline phosphatase. It is a heterogeneous disorder that can have variable disease severity. Adults can present with osteomalacia, chondrocalcinosis, and/or stress fractures. At least one case report documents atypical femoral fractures in a bisphosphonate-treated adult hypophosphatasia patient , highlighting the potential for medication risks. A few case reports have documented efficacy of teriparatide in hypophosphatasia . Enzyme treatment (asfotase alfa) is FDA approved for treatment of hypophosphatasia, but there are currently no guidelines for treatment in adults .
General measures to improve bone health in the context of a need for glucocorticoid therapy include effort to minimize dose and duration of glucocorticoid therapy as well as calcium and vitamin D supplementation . Glucocorticoids may interfere with vitamin D absorption, and so patients requiring glucocorticoid treatment may need higher doses of vitamin D than patients not taking steroids . Patients with prolonged glucocorticoid exposure may have hypogonadism and menstrual irregularities, and in these cases estrogen replacement may also be considered for initial management.
The 2017 American College of Rheumatology Guideline for the Prevention and Treatment of Glucocorticoid Induced Osteoporosis recommends risk stratification of patients based on history of fractures, age, amount and duration of glucocorticoid use. These guidelines recommend consideration of treatment for premenopausal women with GIOP who are at moderate to high fracture risk. This includes women with prior osteoporotic fracture(s) or those with very low BMD (Z-score<−3) or very rapid bone loss and continuing glucocorticoid treatment at ≥7.5 mg of prednisone or equivalent per day for ≥6 months . If pharmacotherapy is merited, oral bisphosphonates or teriparatide are recommended, and longer acting medications are considered only in special circumstances . These very conservative treatment criteria are meant to provide guidance, but do not preclude consideration of medical intervention for less severe bone density findings, if clinically indicated.
As discussed previously, alendronate and risedronate are two bisphosphonates that are FDA approved for the treatment of premenopausal women receiving glucocorticoids. However, few premenopausal women were included in the trials of bisphosphonates for GIOP. Since none of the premenopausal patients in these studies experienced fractures , effects of bisphosphonates on fracture risk in premenopausal GIOP remain uncertain.
Teriparatide has also been examined in patients with GIOP. In one 18-month randomized double-blind trial comparing teriparatide to alendronate in over 400 men and women with osteoporosis (ages 22–89 years) and glucocorticoid exposure ≥5 mg/day prednisone equivalent for ≥3 months, teriparatide 20 μg daily increased BMD and reduced vertebral fracture incidence to a greater extent than alendronate . In a prespecified analysis comparing effects according to gender and menopausal status, the 62 premenopausal women in this study had greater BMD gains at the lumbar spine with teriparatide compared to alendronate (7.0% vs 0.7%, P <.001), which was similar to findings in men and postmenopausal women. However, there did not appear to be a difference in fracture rates between the two premenopausal treatment groups . No premenopausal patients in either treatment group had new vertebral fractures (as was the case with the bisphosphonate trials noted earlier ), and there was no significant difference in the number of patients with nonvertebral fractures.
Premenopausal osteoporosis is rare, but can cause significant morbidity. Women presenting with unexplained fractures or low BMD should have a comprehensive clinical and laboratory evaluation to search for secondary causes of bone fragility and/or bone loss. Initial management should focus on treatment of the underlying cause, when possible. Pharmacologic interventions may be considered for some premenopausal women, particularly those with an ongoing cause of bone loss and those who have had or continue to have major low-trauma fractures. While medications such as bisphosphonates and teriparatide have been shown to improve BMD in several types of premenopausal osteoporosis, studies of this rare condition are small and have not provided evidence of fracture risk reduction with any therapeutic approach.