Reprinted from Eating Disorders Review
November/December Volume 26, Number 6
©2015 iaedp

Bone loss is a common and potentially devastating medical complication of anorexia nervosa (AN). At least half of young women with AN will develop osteoporosis, and their bone loss often occurs relatively early in the disease. It also may not be fully reversible, at least with currently available treatments. It is important to remember that we may not yet appreciate the full impact of osteoporosis in AN. Given the increased prevalence of AN in the last 40 to 50 years, the majority of people who have – or did have – AN may not have yet reached the age when osteoporosis complications are most common and problematic. For this reason, gaining more knowledge about preventing and treating osteoporosis in AN is critically important.

1. Anorexia and Bone Health: A Complex Relationship

In AN patients, bone resorption increases without a corresponding increase in bone formation, leading to a net loss of bone. Levels of bone resorption markers such as N-terminal telopeptide (NTX) and deoxypyridoline may be elevated, but bone formation markers such as osteocalcin are not (Eur J Clin Nutr. 2004:1257. Because significant bone loss can occur so early in AN, even within the first year, early diagnosis and intervention are important.

Baseline screening is recommended for all patients who have had AN or amenorrhea for 6 to 12 months or longer. The Osteoporosis Foundation recommends screening all women under age 65 who have low body weight (Geriatrics. 2000 May; 55:31), and the American College of Sports Medicine recommends screening for osteoporosis in athletes who have a history of low estrogen levels or disordered eating for 6 months or more, or those with a history of stress fracture or any fracture from slight trauma (Med Sci Sports Exerc. 2004;36:1985; Med Sci Sports Exerc. 2007; 39:1867). 

2. Peak Bone Mass Forms between 11 and 14 Years of Age

While peak bone mass is reached by the third decade of life, from 40% to 60% of a woman’s bone mass is gained during the pubertal bone surge from ages 11 through 14, also the time when symptoms of AN often appear. The predictive relationship between bone mineral density (BMD) and short-term fracture incidence is unclear in premenopausal women. For these reasons, the International Society for Clinical Densitometry recommends against the use of T-scores to categorize BMD measurements in premenopausal women. Instead, the Society recommends using Z-scores, which compare women to an age-matched reference population. Young women with BMD Z-scores below — 2.0 should be categorized as having BMD that is “below expected range for age,” and those with Z-scores above — 2.0 should be categorized as having BMD that is “within the expected range for age” (www.iscd.org). 

Drs. Madhusmita Misra and Anne Klibanski at Harvard Medical School point out that as many as 50% of adolescent girls with AN have Z-scores less than -1 recorded at one site and 11% have Z-scores less than -2 recorded at a minimum of one site (Clin Obstet Gynecol. 2013; 56:722). There are no longitudinal prospective studies relating BMD by dual-energy x-ray absorptiometry (DXA) scans to fractures in premenopausal women.

Bone tissue is always in a state of change; old bone is reabsorbed and new bone laid down by osteocytes. During childhood and early adolescence, the osteocytes lay down new bone cells faster than they are being absorbed, leading to peak bone mass. This continues to about age 30, when bone production slows down and bone mass shrinks by about 1% each year afterward.

As Dr. Kathryn Teng of the Cleveland Clinic has written, it is essential to optimize bone mass during these years because even small difference in BMD can have significant implications later. She notes that just a 5% increase in bone density significantly decreases fracture risk, whereas a 10% decrease in adult BMD is associated with twice to three times the risk of fracture (CCJM. 2011; 78:50). After the third decade of life, skeletal mass begins to decrease and is accelerated after the menopause, where the rate of bone loss exceeds that of bone formation. Failure to attain sufficient bone density during adolescence leads to osteopenia, or a bone density between 1 and 2.5 standard deviations below average on Z scores for young adults of the same gender, and is a risk factor for osteoporosis. 

3. Young Patients with Prolonged Amenorrhea Have More Severe and Longer-Lasting Bone Defects

Younger AN patients are at greater risk because often they haven’t reached their peak maximum peak bone mass. Beyond the low DXA measures of BMD, teens with AN have weakened bone microarchitecture. For example, the volume of trabecular (spongy) bone and trabecular thickness are lower and trabecular separation greater in girls with AN than in healthy controls, even when the BMD measurements are the same. This breakdown and weakening of bone microarchitecture is of concern because of the increased risk of fracture risk independent of BMD. 

Young women with amenorrhea should be evaluated to determine if the amenorrhea is primary (absence of menarche by age 16) or secondary (absence of menses for more than 3 cycles or a past history of amenorrhea for more than 6 months). In a study of 73 women with AN, having the illness for 20 months was an important cutoff point: after this, the most severe form of osteopenia was seen (J Endocrinol Metab. 2009; 95:470).Weight gain and resumption of menses led to a 3% annual gain in BMD at the spine and a 2% gain at the hip, compared with a decrease in BMD of about 2.5% at both sites in persons who did not recover. As Drs. Misra and Klibanski point out, however, residual deficits can persist and make it impossible for patients to catch up to normal bone density.

4. Hormonal Changes Also Impact Bone Loss

AN patients also have important endocrine changes in hormonal axes that can affect bone loss. These include the hypothalamic-pituitary-gonadal axis, the growth hormone insulin-like growth factor (IGF-1) axis, and the hypothalamic-pituitary-adrenal axis. (J Clin Endocrinol Metab. 2004; 89: 4920). Many familiar hormones impacted by the patient’s energy status, including peptide YY (PYY), leptin, ghrelin, insulin and adiponectin are affected, and in turn may affect bone health. 

Teens with AN have higher-than-normal serum and urinary cortisol levels, and increased cortisol is associated with a lower body mass index and fat mass and lower glucose levels. According to Drs. Misra and Klibanski, this is an adaptive response to malnutrition, as cortisol is gluconeogenic. But, higher cortisol levels are also associated with lower BMD and inversely with bone turnover markers (J Clin Endocrinol Metab. 2004; 89: 4920). These higher cortisol levels are predictive of lower spine and hip BMD, as well as lower extremity lean mass in AN, which may also impact bone health (J Clin Endocrinol Metab. 2005 95;2580). The hypoestrogenic state noted in AN, a direct result of the effects of hypothalamic amenorrhea, is further worsened by changes in estrogen metabolism.

5. Weight Restoration is the Cornerstone of Treatment

According to Dr. Teng and others, weight restoration is the cornerstone of treatment for low BMD due to AN. All AN patients should be referred to a nutritionist to develop a meal plan. Treatment goals for patients with disordered eating are designed to optimize the patient’s nutritional status, normalize her or his eating behavior, change unhealthy thoughts that maintain the disorder and treat emotional issues that are contributing to the disordered eating.

Restoring weight and with it menses are two of the most important goals of treatment because young women with AN can lose as much as 2% to 6% of their bone mass each year they have AN. While previous studies have shown that BMD increases as the patient recovers, long-term follow-up studies show that loss of bone density may not be completely reversible, even when menses resumes and the patient regains weight. 

6. Estrogen Replacement Is Ineffective

Unlike the positive effects among postmenopausal women, where estrogen therapy maintains or improves BMD and may help reduce the incidence of fractures, the same pattern has not been seen in premenopausal women with osteoporosis. Estrogen Replacement seems to have little independent effect on correcting or preventing osteopenia in AN (Eur J Endocrinol. 2002; 146:45). Some researchers report that this is due to the possibility that the natural cycle of estrogen may be important to bone recovery. In postmenopausal women, estrogen therapy appears to work by impairing osteoclast-mediated bone resorption, but it has only slight effects on bone formation. In premenopausal women with anorexia, in contrast, bone loss appears to be due to a unique “uncoupling” of osteoblastic and osteoclastic functions, according to Dr. Teng and others (Int J Eat Disord. 2001; 29:11). This uncoupling of osteoblastic and osteoclastic functions results in reduced bone formation and increased bone resorption—estrogen may not improve this.

The American College of Sports Medicine recommends considering estrogen therapy if there is evidence of a decline in BMD in an athlete older than 16 years of age who also has persistent functional hypothalamic amenorrhea despite adequate nutritional intake and weight regain (Br J Sports Med. 2014; 48:289). However, the Society also acknowledges that restoring regular menstrual cycles with oral contraceptives will not normalize metabolic factors that impair bone formation, health and performance, and again note that it is unlikely that estrogen will fully reverse low BMD in these patients.

7. Bisphosphonates Are Not Approved for Premenopausal Patients with AN

The FDA has approved use of bisphosphonates only for premenopausal women who are also taking glucocorticoids. The bisphosphonates do increase BMD in young women with AN but at the same time are teratogenic and have a long half-life that enables the drugs to affect bone turnover for up to 2 years after they are stopped (Am J Med. 1997; 103:291). For patients who purge via vomiting, the bisphosphonates also increase the risk of esophageal ulcerations. Thus, for the time being, the bisphosphonates do not represent a major treatment option for osteoporosis in patients with AN.

8. Calcium Supplements Do Not Increase BMD

The National Osteoporosis Association has recently issued a statement in response to the discussion about the need (or not) for calcium supplements. Noting that calcium is needed for strengthening collagen in bone tissue and for many other functions, the organization recommends attempting to get “the recommended daily amount of calcium from food first, and to supplement this as needed to make up for the shortfall.” (For the Association’s full statement, see www.osteoporosis.org .) The organization also mentioned that calcium and vitamin D were recently reaffirmed by the 2015 U.S. Dietary Guidelines Advisory Committee as nutrients of public health concern because their under-consumption has been linked to adverse health outcomes. (http://health.gov/dietaryguidelines/2015-scientific-report/PDFs/Scientific-Report-of-the-2015-Dietary-Guidelines-Advisory-Committee.pdf). 

9. Always Monitor Exercise, Particularly In Patients Who Are Also Athletes

Strenuous exercise is a major factor in maintaining and developing normal bone mass; however, the benefit of exercise on bone loss in anorectic patients is still a topic of debate. The amount of exercise has to be monitored to help limit further weight loss. In addition, abnormal plasma estradiol concentrations may limit the positive effects of exercise on bone density (Calcif Tissue Int. 2000; 67:277). 

— MKS