Joint Bone Spine 72 (2005) 489495 http://france.elsevier.

com/direct/BONSOI/

Review

Bone loss associated with anorexia nervosa

Isabelle Legroux-Gerot a,*, Jean Vignau b, Francis Collier c, Bernard Cortet a
Service de Rhumatologie, CHRU Lille, Hpital Roger Salengro, 59037 Lille cedex, France b Service dAddictologie, CHRU Lille, Hpital La Charit, 59037 Lille cedex, France c Service de Gyncologie, CHRU Lille, Hpital Jeanne de Flandre, 59037 Lille cedex, France Received 2 February 2004; accepted 15 July 2004 Available online 05 October 2004
a

Abstract The objective of this study was to evaluate the epidemiology, diagnosis, pathophysiology, and treatment of bone loss related to anorexia nervosa. Earlier onset and longer duration of anorexia nervosa are associated with more severe bone loss. Osteoporosis develops in 3850% of cases. Bone mineral density measurement by dual-energy X-ray absorptiometry is useful for assessing bone mass, and bone marker assays provide information on bone turnover. Bone loss in anorexia nervosa is probably multifactorial. Estrogen deciency was long felt to be the major factor. However, in contrast to postmenopausal osteoporosis, bone loss associated with anorexia nervosa is related mainly to inadequate bone formation, with only a slight increase in bone resorption. This suggests a role for nutritional factors, such as disturbances in the growth hormonesomatomedin C axis (GH/IGF-I) related to malnutrition. The best treatment strategy for correcting bone mass in patients with anorexia nervosa is not agreed on. Resumption of menstrual cycles and weight gain seem necessary but not always sufcient. Studies found no benets with estrogen therapy, but this was usually given as estrogenprogestin contraceptives. No vast studies evaluating hormone replacement therapy have been reported. Bone formation enhancers such as IGF-I seem to provide the best results, most notably when used in combination with estrogens. This suggests that complex treatment strategies combining bone formation enhancers and bone resorption inhibitors may deserve evaluation. 2004 Published by Elsevier SAS.
Keywords: Anorexia nervosa; Osteoporosis; IGF-I

Anorexia nervosa has become a major public health problem over the last few years in industrialized countries. Its prevalence is 0.5%, as compared to 2% for bulimia [1,2]. Major morbidity occurs in association with anorexia nervosa. Bone loss is of special concern. Studies found bone mass decreases at the spine or femoral neck greater than 2.5 standard deviations (S.D.) in 3850% of patients and greater than 1 S.D. in 92% of patients [35]. Multiple hormonal, endocrine, and nutritional factors affect bone metabolism in patients with anorexia nervosa (Fig. 1) [1]. Anorexia nervosa occurs during adolescence, which is a critical period for peak bone mass achievement. The slow steady bone mass increase that occurs throughout childhood gives way to a spurt during adolescence, the rate being fastest at Tanners stage
* Corresponding author. Tel.: +33 3 20 44 44 15; fax: +33 3 20 44 54 62. E-mail address: i-legroux@chru-lille.fr (I. Legroux-Gerot). 1297-319X/$ - see front matter 2004 Published by Elsevier SAS. doi:10.1016/j.jbspin.2004.07.011

4 or 5. Genetic factors may explain 6080% of the peak bone mass value; nutritional and hormonal factors account for the remaining 2040% [6]. Bone loss is more severe in women with anorexia nervosa than in same-age women with a normal body mass index (BMI) and hypothalamic amenorrhea. In one study, criteria for osteoporosis were met in 40% of women with anorexia nervosa as compared to only 16% of those with hypothalamic amenorrhea [7]. Bone mineral density (BMD) is signicantly lower in women who are younger than 18 years at onset of anorexia than in those older than 18 years, attesting to the impact of anorexia nervosa on bone formation [8]. The objective of this article is to review the epidemiology, diagnosis, pathophysiology, and treatment of bone loss associated with anorexia nervosa. Anorexia nervosa is dened as a combination of exaggerated concern about gaining weight, alterations in body image, major weight loss, refusal to maintain body weight within the normal range, and amenorrhea [9].

490

I. Legroux-Gerot et al. / Joint Bone Spine 72 (2005) 489495

Fig. 1. Pathophysiology of bone loss in patients with anorexia nervosa.

1. Evaluation of bone abnormalities in patients with anorexia nervosa 1.1. Bone mineral density Bone mass can be measured by absorptiometry at the spine and femoral neck. Radiation exposure is minimal with this method. The World Health Organization denes osteoporosis in postmenopausal women as a BMD value at least 2.5 S.Ds. below the mean in young women (T-score < 2.5 S.Ds.), at the spine, femoral neck or radius. This denition may be unsatisfactory in adolescents, who may not have achieved their peak bone mass. Studies comparing BMD values at various sites in patients with anorexia nervosa and in healthy controls consistently found lower values in the patients, with an about 50% rate of osteoporosis [35,10 12]. 1.2. Markers for bone turnover Several markers are available for assessing bone turnover. Although they are useful complements to BMD measurement, they cannot provide a diagnosis. The most widely used bone formation markers are osteocalcin and bone alkaline phosphatase. Bone resorption markers include deoxypyridinoline, C-terminal peptides of type 1 collagen (cross-laps or CTX), N-terminal peptides of type 1 collagen (NTX), and procollagen type 1 carboxy-terminal extension peptide (ICTP) [11]. These markers are used chiey in postmenopausal women and may be difcult to interpret in young women and adolescents. Whereas increased bone resorption is a feature shared by anorexia nervosa and postmenopausal bone loss, a marked decrease in bone formation is also present in patients with anorexia nervosa [1315], indicating a role for nutrition or nutrition-dependent factors in addition to estrogen deprivation. Further support for this multifactorial mechanism comes from a study showing that bone loss is more severe in patients with anorexia nervosa than in same-age women with hypogonadism [7].

Few studies have evaluated the risk of fractures in populations with anorexia nervosa [3,1618]. In a retrospective study of 208 patients, Lucas et al. [19] recorded 58 fractures over a 13-year period, a three-fold increase over the expected number. Fractures were more common in the patients who received inpatient care, and bone insufciency ssures were also prevalent. In another study, the fracture risk was increased seven-fold in patients with a mean disease duration of 5.8 years, as compared to same-age healthy controls [20]. The distribution of fractures was the same as in osteoporosis, with the most common sites being the spine, radius, and proximal femur, in that order.

2. Mechanisms underlying bone loss in patients with anorexia nervosa 2.1. Hormonal factors A few studies investigated BMD changes in patients with anorexia nervosa [5,11,12,20,21]. The BMD values were signicantly lower in the patients who were younger than 18 years at disease onset, illustrating the impact of anorexia nervosa on peak bone mass achievement. Amenorrhea is a diagnostic criterion for anorexia nervosa, and estrogen deciency has been described as a major source of bone loss in this condition. The mechanisms underlying this estrogen deciency remain incompletely understood but involve multiple factors including hypothalamic dysfunction, weight loss, and dysregulation of neurotransmitters such as GnRH. The severity of bone loss was correlated with age at onset and disease duration in several studies [5,11,20,22]. Osteoporosis has been reported also in adults with hypogonadism (Turner syndrome, delayed puberty, aromatase deciency), hyperprolactinemia, or hypothalamic hypogonadism. Bone loss is less severe in these diseases and in ovariectomized patients, however, than in patients with anorexia nervosa. These data indicate a role for factors other than estrogen deciency. Further

I. Legroux-Gerot et al. / Joint Bone Spine 72 (2005) 489495

491

evidence that bone loss associated with anorexia nervosa is not entirely ascribable to estrogen deciency comes from studies of bone turnover patterns. Postmenopausal osteoporosis is associated with increases in both bone resorption and bone formation with, however, a greater increase in resorption, so that the net result is bone loss. In contrast, abnormalities in bone formation predominate in patients with anorexia nervosa: bone resorption is slightly increased but bone formation is either normal [5] or decreased [7,11]. This predominant abnormality in osteoblastic function explains the limited effectiveness of estrogens and other antiresorptive agents [15]. Thus, malnutrition and nutrition-dependent factors play a key role in the pathogenesis of bone loss associated with anorexia nervosa. 2.2. Nutritional and endocrine factors Strong correlations have been reported between BMD values in patients with anorexia nervosa and variables reecting nutritional status, such as BMI, fat-free mass, body fat, IGF-I, and leptin [1,7,11,23,24]. Hotta et al. [25] reported a high risk of osteoporosis in patients, whose BMI was lower than 15 kg/m2. Other studies [11,25,26] found that bone formation markers (osteocalcin and bone alkaline phosphatase) were positively correlated with markers for nutritional status such as BMI, percent body fat, and IGF-I, whereas bone resorption markers were negatively correlated with estradiol levels. Audi et al. [26] studied the effect of disease activity on bone mass and bone turnover markers in 73 patients with anorexia nervosa and Tanner 5 pubertal development. The patients were divided into three groups, as follows: full-blown disease, partial weight regain but persistent amenorrhea, and full weight regain with normal menstrual cycles for a mean of 10 6 months. Mean age was similar in these three groups (17 1.6 years, 17.3 1.8 years, and 17.6 1.8 years, respectively). Lumbar spine BMD was not signicantly different across the three groups but osteopenia was less common among the patients who had recovered, as compared to the other two groups. IGF-I and estradiol levels were signicantly lower in the group with full-blown disease, whereas bone formation markers (osteocalcin, bone alkaline phosphatase, procollagen type 1 carboxy-terminal propeptide, and procollagen type 1 amino-terminal propeptide) were signicantly higher in the partly recovered and fully recovered groups. Among markers for bone resorption (ICTP, deoxypyridinolines, and N-terminal peptides of collagen 1), only ICTP showed signicant differences, with lower levels in the partly recovered patients than in those with full-blown disease, as well as lower levels in the fully recovered patients than in the partly recovered patients. In this study, BMD was positively correlated with body weight, height, and time with menstrual cycles prior to amenorrhea onset in the patients with fullblown disease or partial recovery. Among the fully recovered patients, as many as 36% had osteoporosis. In this group, the only difference between patients with and without osteoporosis was the pattern of bone turnover markers: osteoporosis

was associated with higher levels of bone formation markers (osteocalcin and procollagen type 1 amino-terminal propeptide) and bone resorption markers (ICTP). Soyka et al [12] studied 19 patients with anorexia nervosa and a bone age ranging from 13 to 18 years, comparing them to 19 controls. Mean disease duration was 19 5 months. Lumbar spine BMD was signicantly lower in the patients with anorexia nervosa, and bone loss severity was correlated with disease and amenorrhea duration. Serum estradiol and IGF-I levels were severely reduced in the patients with anorexia. Bone formation markers (osteocalcin and bone alkaline phosphatase) were signicantly decreased, whereas bone resorption markers (deoxypyridinoline and N-terminal peptides of collagen 1) were unchanged as compared to the controls. IGF-I levels were closely correlated with bone formation markers, indicating abnormal osteoblast function in the patients with anorexia nervosa. Abnormalities in the growth hormone (GH)-insulinlike growth factor I (IGF-I) axis have been described in patients with anorexia nervosa [1,2729]. These hormones increase during puberty and stimulate the proliferation and differentiation of osteoblast precursors. IGF-I enhances bone formation and growth via effects on osteoblasts and collagen synthesis. Resistance to GH with high GH levels but low IGF-I levels has been found in patients with anorexia nervosa. In a study of GH production over the 24-hour cycle in 8 patients with anorexia nervosa, Stoving et al. [27] found increases in peak number, duration, and amplitude, as well as a 20-fold increase in basal secretion and a four-fold increase in pulsatile secretion. The increased peak amplitude may be related to weight loss and the increased peak number to estrogen deciency. The half-life of GH was similar in the patients and the healthy controls. Sacchi et al. [29] reported similar ndings. Several studies found decreases not only in IGF-I levels, but also in levels of the IGF-I binding proteins (IGFBD) 2 and 3, in some cases with an increase in IGFBP1 [3035]. The decrease in circulating IGFBP levels may contribute to generate resistance to GH by preventing IGF-I transport to target organs. Furthermore, IGFBP3 may be a strong predictor of bone loss in patients with anorexia nervosa, independently from BMI and IGF-I levels [35]. Argente et al. [32] reported that weight regain was associated with a return to normal of GH but not of IGF-I or IGFBP3. Thus, in patients with anorexia nervosa, the peripheral axis was independent from GH and the GH/GFI-I axis remained abnormal despite weight regain. In sum, bone loss associated with anorexia nervosa is due to a complex set of hormonal, endocrine, and nutritional factors. The respective contributions of hypercorticism, calcium and vitamin D deciency, and excessive physical activity in some patients remain to be determined. High cortisol levels with a normal circadian cycle have been reported [1,8,26]. High urinary excretion of free cortisol with hypercorticism suppression by dexamethasone is common. Hypothalamic dysfunction or CRH overproduction may contribute to the hypercorticism. Grinspoon et al. [5] reported hypercorticism in only 22% of patients with anorexia nervosa and severe bone

492

I. Legroux-Gerot et al. / Joint Bone Spine 72 (2005) 489495

loss, and Audi et al. found that urinary excretion of free cortisol was similar in patients with anorexia nervosa and in controls. These data indicate that hypercorticism may contribute to bone loss in some patients but is not the only factor involved. The role for vitamin D and calcium deciency remains unclear. Audi et al. [26] found low levels of 25OHD3 in 24.6% of patients with anorexia nervosa. Urinary calcium excretion was high in patients with full-blown disease and low in those with a partial or complete recovery. Soyka et al. [12] noted an inadequate dietary calcium intake (<1300 mg/d) in 42% of patients with anorexia nervosa but also in 50% of controls. Similarly, vitamin D deciency was present in 42% of the patients and 44% of the controls.

4.1.2. Physical activity Conicting results have been reported [40]. Although physical activity is needed to achieve peak bone mass and to maintain bone stock in adulthood, its ability to protect against osteoporosis in patients with anorexia nervosa is a matter of controversy. Physical activity was associated with an increase in cortical BMD values in some studies [3,41] but in others had no effect, regardless of the level of exertion [12,24]. A few studies established that strenuous physical activity in elite athletes caused bone tissue alterations when the activity level was sufcient to cause weight loss and amenorrhea [42,43]. 4.2. Pharmacotherapy 4.2.1. Calcium and vitamin D supplementation Although calcium and vitamin D supplementation makes good sense, most notably in patients with deciencies in these compounds, effectiveness has been inconsistent. In one study, bone loss was reduced in patients given 1500 mg of calcium per day [44], whereas in another study calcium supplementation had no effect [45]. Abrams et al. [46] reported alterations in calcium metabolism with decreased intestinal absorption and increased urinary excretion. 4.2.2. Estrogens Despite the documented association between anorexia nervosa and estrogen deciency and the strong correlation between bone loss and amenorrhea duration, most studies found no improvement in bone mass with estrogenprogestin therapy (Table 1). It should be noted that estrogens were usually supplied as estrogenprogestin contraceptives. Seeman et al. [47] reported an increase in lumbar BMD values after 30 months on an estrogenprogestin contraceptive, but the values remained substantially lower than in the control group, and no effect was detected at the femoral neck. Klibanski et al. [48] conducted a randomized placebo-controlled study of oral estrogenprogestin therapy with calcium supplementation in 48 patients with anorexia nervosa. Among the treated patients (n = 22), 16 received hormone replacement therapy with conjugated equine estrogens (PremarinProvera) and 6 an oral contraceptive (35 g ethinyl estradiol per day). After 18 months, bone mass was similar in the treated patients and in the controls. However, treatment efcacy was strongly dependent on body weight at baseline, with the largest bone gains being seen in the patients with the lowest baseline weights expressed as the percentage of the ideal weight. Among patients whose baseline body weight was less than

3. Changes in bone mass after weight regain The BMD changes during recovery from anorexia nervosa have been investigated [3,10,12,16,20,22,24]. Although bone mass increased when weight returned to normal, several studies showed persistent osteopenia in a large proportion of patients [3,10,12,16,36]. Hartman et al. [37] studied bone mass at in 19 women with a history of anorexia nervosa followed by a full recovery for a mean of 21 years. Although their body weight was normal, their bone mass at the femoral neck was low as compared to the control group. Serial absorptiometry measurements at the spine allowed Zipfel et al. [20] to document a gain in bone mass after weight regain with decreases in the percentages of patients with osteopenia (from 35% to 13%) and osteoporosis (from 54% to 21%). Nevertheless, BMD values remained low in many patients. Wentz et al. [38] failed to replicate these ndings in a recent study, in which BMD values in patients with a history of anorexia nervosa and a mean time since recovery of 11 years were similar to those in controls.

4. Treatments 4.1. Nonpharmacological treatments 4.1.1. Weight regain and menstrual cycle recovery These two factors improved bone mass in several studies [7,10,14,22,39] but were not sufcient in another study [12]. Although weight regain and menstrual cycle recovery seem to be prerequisites to bone mass gain, they do not allow a return to normal bone mass values.

Table 1 Published data on the effectiveness of estrogen therapy in patients with anorexia nervosa Authors Seeman et al. [47] Klibanski et al. [48] Golden et al. [49] Patel [50] Patients 16 treated vs. 49 22 treated vs. 26 22 treated vs. 28 1 patient Treatments OC Follow-up (months) 31.8 months BMD changes Spinal BMD (1.4 0.005 vs. 1.02 0.002; P < 0.02). No change at the femoral neck No difference with treatment No difference with treatment Spinal BMD by 38%. No change at the femoral neck.

16 on CEEs 6 on OC 18 months OC 23 months Estradiol 2 mg 12 months

BMD, bone mineral density; CEEs, conjugated equine estrogens; OC, oral contraceptive.

I. Legroux-Gerot et al. / Joint Bone Spine 72 (2005) 489495

493

70% of their ideal weight, 10 were controls and experienced a 20.1% 16.2% drop in their mean BMD value and six were treated patients and had a 4% 8.8% increase in mean BMD value. No signicant BMD differences were found between the two groups when only those patients whose baseline body weight was greater than 70% of their ideal weight were considered. No signicant differences were noted between the patients given PremarinProvera and those given an oral contraceptive. It should be pointed out that BMD measurements were done by quantitative computed tomography in this study rather than by absorptiometry. Golden et al. [49] investigated bone mass changes after 23 months of oral contraceptive therapy in 50 patients with anorexia nervosa and a mean age of 16.8 years. No signicant increases in bone mass were found at the spine or femoral neck after 1 year, although the patients gained weight. A single study [50] found an increase in bone mass after estrogen therapy; however, the bone gain was moderate and the patients had severe osteoporosis at baseline. Thus, estrogen deciency does not fully explain the loss of bone mass associated with anorexia nervosa. There is loss of the normal coupling of bone formation and resorption: bone resorption is increased but bone formation is not. This may explain the inefcacy of estrogen therapy, which diminishes bone resorption by reducing the production of osteoclastactivating cytokines (IL-1, IL-6, TNFa, and PGE2) and by increasing the production of factors that inhibit osteoclast differentiation and activation (TGFb and osteoprotegerin). Estrogens have only limited effects on bone formation, and the main mechanism of bone loss in anorexia nervosa is loss of coupling between an abnormally high bone resorption rate and an unchanged bone formation rate. Many hypotheses have been put forward to explain the failure of estrogen therapy to improve bone mass in patients with anorexia nervosa [15,51]. A low dietary intake of carbohydrates may alter estrogen metabolism. In addition, a low calorie intake directs the metabolism of estrogens toward conversion into metabolites that have weak estrogen effects [51]. Finally, poor compliance with medical prescriptions and poor tolerance of estrogens are common in patients with anorexia nervosa and may lead to use of insufcient dosages. 4.2.3. IGF-I Several studies found that patients with anorexia nervosa had low levels of IGF-I, a hormone that contributes to bone growth by stimulating the osteoblasts [1,7,11,23,24,27]. This prompted interest in potential benets of IGF-I therapy on bone mass in patients with anorexia nervosa. Grinspoon et al. [34] allocated 60 patients at random to IGF-I alone (30 g/kg bid subcutaneously, IGF-I plus oral contraception, oral contraception alone, or a placebo. Calcium and vitamin D supplements were given to patients in all four groups. Follow-up was 9 months. Total bone mass increased signicantly in the groups given IGF-I as compared to the placebo (1.1% 0.5% vs. 0.6% 0.8%; P = 0.05). A signicant increase in spinal BMD vs. the placebo group was noted only in the group given

Fig. 2. Changes in lumbar spine BMD in four groups of patients with anorexia nervosa treated for 4 months: Group I, IGF-I + OC; Group II, IGF-I alone; Group III, OC alone; Group IV, placebo. From Grinspoon et al. [34]. OC, oral contraception (estrogen-progestin). * P < 0.05 vs. controls (Group IV).

both IGF-I and oral contraception (1.8% 0.8% vs. 1% 1.3%; P = 0.05) (Fig. 2). At the other sites, however, IGF-I plus oral contraception (or IGF-I alone) failed to produce signicant BMD increases vs. the placebo. At the hip, changes in BMD (g/cm2) in the four groups were as follows: IGF-I plus oral contraception, 0.008 0.007; IGF-I alone, 0.007 0.010; oral contraception alone, 0.003 0.008; and placebo, 0.004 0.005. 4.2.4. Bisphosphonates Bisphosphonate therapy does not seem advisable in adolescents. Their predominant antiresorptive effect argues against their use in patients with anorexia nervosa. No data are available on bisphosphonate therapy in this setting. Nevertheless, combining a bisphosphonate with a bone anabolism enhancer might deserve consideration. 4.2.5. Parathyroid hormone Parathyroid hormone therapy has not been investigated in patients with anorexia nervosa. The effect of parathyroid hormone on bone formation may hold promise in the treatment of bone loss associated with anorexia nervosa. However, the risk of osteosarcoma may limit the usefulness of parathyroid hormone in these patients, who are young and often in a phase of rapid skeletal growth.

5. Conclusion Anorexia nervosa is associated with severe bone loss that must be looked for routinely by obtaining absorptiometry measurements of BMD. The mechanism is multifactorial and incompletely understood. Diminished bone formation may play a greater role than increased bone resorption. The best management is not agreed on. Oral contraceptives have been

494

I. Legroux-Gerot et al. / Joint Bone Spine 72 (2005) 489495 [20] Zipfel S, Seibel MJ, Lowe B, Beumont PJ, Kasperk C, Herzog W. Osteoporosis in eating disorders: a follow-up study of patients with anorexia and bulimia nervosa. J Clin Endocrinol Metab 2001;86: 522733. [21] Seeman E, Kerlsson MK, Duan Y. On exposure to anorexia nervosa, the temporal variation in axial and appendicular skeletal development predisposes to site-specic decits in bone size and density: a crosssectional study. J Bone Miner Res 2000;15:225965. [22] Valla A, Groenning IL, Syversen U, Hoeiseth A. Anorexia nervosa: slow regain of bone mass. Osteoporos Int 2000;11:1415. [23] Jacoangeli F, Zoli A, Taranto A, Staar Mezzasalma F, Ficoneri C, Pierangeli S, et al. Osteoporosis and anorexia nervosa: relative role of endocrine alterations and malnutrition. Eat Weight Disord 2002;7: 1905. [24] Bachrach LK, Guido D, Katzman D, Litt IF, Marcus R. Decreased bone density in adolescent girls with anorexia nervosa. Pediatrics 1990;86:4407. [25] Hotta M, Shibasaki T, Sato K, Demura H. The importance of body weight history in the occurrence and recovery of osteoporosis in patients with anorexia nervosa: evaluation by dual X-ray absorptiometry and bone metabolic markers. Eur J Endoc 1998;139:27683. [26] Audi L, Vargas DM, Gussiny M, Yeste D, Marti G, Carrascosa A. Clinical and biochemical determinants of bone metabolism and bone mass in adolescent female patients with anorexia nervosa. Pediatric Res 2002;51:497504. [27] Stoving RK, Veldhuis JD, Flyvbjerg A, Vinten J, Hangaard J, Koldkjaer OG, et al. Jointly amplied basal and pulsatile Growth Hormone (GH) secretion and increased process irregularity in women with anorexia nervosa: indirect evidence for disruption of feedback regulation within the GH-Insulin-Like Growth Factor I axis. J Clin Endocrinol Metab 1999;:205662. [28] Nussbaum MP, Blethen SL, Chasalow FI, Jacobson MS, Shenker IR, Feldman J. Blunted growth hormone responses to clonidine in adolescent girls with early anorexia nervosa. Evidence for an early hypothalamic defect. J Adolesc Health Care 1990;11:1458. [29] Scacchi M, Pincelli AI, Caumo A, Tomasi P, Delitala G, Baldi G, et al. Spontaneous nocturnal growth hormone secretion in anorexia nervosa. J Clin Endocrinol Metab 1997;82:32259. [30] Counts DR, Gwirtsman H, Carlsson LMS, Lesem M, Cutler Jr. GB. The effect of anorexia nervosa and refeeding on growth hormonebinding protein, the insulin-like growth factor (IGFs), and the IGFbinding proteins. J Clin Endocrinol Metab 1992;75:7627. [31] Golden NH, Kreitzer P, Jacobson MS, Chasalow FI, Schebendach J, Freedman SM, et al. Disturbances in growth hormone secretion and action in adolescents with anorexia nervosa. J Pediatr 1994;125:655 60. [32] Argente J, Munoz MT, Pozo J, Barrios V, Buno M, Chowen JA, et al. Growth hormone resistance in anorexia nervosa as a model of malnutrition. J Endocrinol Invest 1998;21:248. [33] Hotta M, Fukuda I, Sato K, Hizuka N, Shibasaki T, Takano K. The relationship between bone turnover and body weight, serum insulinlike growth factor (IGF) I, and serum IGF-binding protein levels in patient with anorexia nervosa. J Clin Endocrinol Metab 2000;85: 2006. [34] Grinspoon S, Thomas L, Miller K, Herzog D, Klibanski A. Effects of recombinant human IGF-1 and oral contraceptive administration on bone density in anorexia nervosa. J Clin Endocrinol Metab 2002;87: 288391. [35] Grinspoon S, Miller K, Herzog D, Clemmons D, Klibanski A. Effects of recombinant human insulin-like growth factor (IGF)-I and estrogen administration on IGF-I, IGF binding protein (IGFBP)-2, and IGFBP3 in anorexia nervosa: a randomised-controlled study. J Clin Endocrinol Metab 2003;88:11429. [36] Kooth SW, Noriega E, Leslie K, Muller C, Harrisson JE. Bone mass and soft tissue composition in adolescents with anorexia nervosa. Bone 1996;19:1818.

of little help. Hormone replacement therapy has not been investigated in large studies. Bone formation enhancers such as IGF-I may produce the best results, most notably when given in combination with estrogens. Thus, drug combinations may hold promise for the treatment of osteoporosis in patients with anorexia nervosa.

References
[1] Munoz MT, Argente J. Anorexia nervosa in female adolescents: endocrine and bone mineral density disturbances. Eur J Endoc 2002;147: 27586. Hsu LK. Epidemiology of eating disorders. Psychiatric Clin North Am 1996;19:681760. Herzog W, Minne H, Deter C, Leiding G, Schellberg D, Wuster C, et al. Outcome of bone mineral density in anorexia nervosa patients 11,7 years after rst admission. J Bone Miner Res 1993;8:597605. Zipfel S, Beumont PJ, Russel J, Herzog W. Osteoporosis in eating disorders. Eur Eat Disord Rev 2000;8:10816. Grinspoon S, Thomas E, Pitts S, Gross E, Mickley D, Miller K, et al. Prevalence and predictive factors for regional osteoporosis in women with anorexia nervosa. Ann Intern Med 2000;133:7904. Heaney RP, Abrams S, Dawson-Hughes B, Looker A, Marcus R, Matkovic V, et al. Peak bone mass. Osteoporos Int 2000;11:9851009. Grinspoon S, Miller K, Coyle C, Krempin J, Armstrong C, Pitts S, et al. Severity of osteopenia in oestrogen-decient women with anorexia nervosa and hypothalamic amenorrhea. J Clin Endocrinol Metab 1999;84:204955. Biller BMK, Saxe V, Herzog DB, Rosenthal DI, Holzman S, Klibanski A. Mechanisms of osteoporosis in adult and adolescents women with anorexia nervosa. J Clin Endocrinol Metab 1989;68:54854. Yager J, Andersen A, Devlin M, Mitchell J, Powers P, Yates A. American Psychiatric Association practice guidelines for eating disorders. Am J Psychiatry 1993;150:20728. Bachrach LK, Katzman DK, Litt IF, Guido D, Marcus R. Recovery from osteopenia in adolescent girls with anorexia nervosa. J Clin Endocrinol Metab 1991;72:6026. Soyka LA, Grinspoon S, Levitsky LL, Herzog DB, Klibanski A. The effects of anorexia nervosa on bone metabolism in female adolescents. J Clin Endocrinol Metab 1999;84:448996. Soyka LA, Misra M, Frenchman A, Miller K, Grinspoon S, Schoenfeld DA, et al. Abnormal bone mineral accrual in adolescent girls with anorexia nervosa. J Clin Endocrinol Metab 2002;87:417785. Lennkh C, De Zwaan M, Bailer U, Strand A, Nagy C, El-Giamal N, et al. Osteopenia in anorexia nervosa; specic mechanisms of bone loss. J Psychiatr Res 1999;33:34956. Stefanis N, Mackintosh C, Abraha HD, Treasure J, Moniz C. Dissociation of bone turnover in anorexia nervosa. Ann Clin Biochem 1998;35:70916. Bolton JGF, Patel S. Osteoporosis in anorexia nervosa. J Psychosom Res 2001;50:1778. Brotman AW, Stern TA. Osteoporosis and pathologic fractures in anorexia nervosa. Am J Psychiatry 1985;142:4956. Maugars Y, Clochon P, Grardel B, Hary S, Phelip X, Forestier R, et al. tude du retentissement osseux dans 33 cas danorexie mentale dont 8 avec une ostoporose fracturaire. Rev Rhum Mal Osteoart 1991;58: 7518. Herzog W, Deter HC, Fiehn W, Petzold E. Medical ndings and predictors of long-term physical outcome in anorexia nervosa: a prospective, 12-year follow-up study. Psychol Med 1997;27:26979. Lucas AR, Melton LJ, Crowson CS, OFallon WM. Long-term fracture risk among women with anorexia nervosa: a population-based cohort study. Mayo Clin Proc 1999;74:9727.

[2] [3]

[4] [5]

[6] [7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15] [16] [17]

[18]

[19]

I. Legroux-Gerot et al. / Joint Bone Spine 72 (2005) 489495 [37] Hartman D, Crisp A, Rooney B, Rackow C, Atkinson R, Patel S. Bone density of women who have recovered from anorexia nervosa. Int J Eat Disord 2000;28:10712. [38] Wentz E, Mellstrm D, Gillberg C, Sundh V, Gillberg IC, Rastam M. Bone density 11 years after anorexia nervosa onset in a controlled study of 39 cases. J Eat Disord 2003;34:3148. [39] Golden NH. Osteopenia and osteoporosis in anorexia nervosa. Adolesc Med 2003;14:97108. [40] Mehler PS. Osteoporosis in anorexia nervosa: prevention and treatment. Int J Eat Disord 2003;33:11326. [41] Rigotti NA, Nussbaum SR, Herzog DB, Neer RM. Osteoporosis in women with anorexia nervosa. N Engl J Med 1984;311:16016. [42] Keen AD, Drinkwater BL. Irreversible bone loss in former amenorrheic athletes. Osteoporos Int 1997;7:3115. [43] Yeager KK, Agostini R, Nattiv A, Drinkwater BL. The female athlete triad: disordered eating, amenorrhea, osteoporosis. Med Sci Sports Exerc 1993;25:7757. [44] Spear BA. Position of the American Diabetic Association: nutrition intervention in the treatment of anorexia nervosa and bulimia. J Am Diabetic Ass 2001;101:8109. [45] Rock CL. Nutritional and medical assessment and management of eating disorders. Nut Clin Care 1999;2:33243.

495

[46] Abrams SA, Silber TJ, Esteban NV, Vieire NE, Stuff JE, Meyers R, et al. Mineral balance and bone turnover in adolescents with anorexia nervosa. J Pediatr 1993;123:32633. [47] Seeman E, Szmukler GI, Formica C, Tsalamandris C, Mestrovic R. Osteoporosis in anorexia nervosa: the inuence of peak bone density, bone loss, oral contraceptive use, and exercise. J Bone Miner Res 1992;:146774. [48] Klibanski A, Biller BM, Schoenfeld DA, Herzog DB, Saxe VC. The effects of estrogen administration on trabecular bone loss in young women with anorexia nervosa. J Clin Endocrinol Metab 1995;80: 898904. [49] Golden NH, Lanzkowsky L, Schebendach J, Palestro CJ, Jacobson MS, Shenker R. The effect of estrogenprogestin treatment on mineral density in anorexia nervosa. J Pediatr Adolesc Gynecol 2002; 15:13543. [50] Patel S. Effects of oestrogen therapy on bone density in a patient with severe osteoporosis and anorexia nervosa. Ann Rheum Dis 1999;1: 5866. [51] Anderson KE, Kappas A, Conney AH, Bradlow HL, Fishman J. The inuence of dietary protein and carbohydrate on the principal oxidative biotransformations of estradiol in normal subjects. J Clin Endocrinol Metab 1984;59:1037.