Anorexia Nervosa
Conditions
Keywords
Anorexia Nervosa, Amenorrhea, Bone Mass, Growth hormone
Brief summary
This study is to determine the effects of anorexia nervosa on bone mass and hormone levels in adolescents. Whether administration of estrogen, a normal hormone present during puberty, can help maintain bone development in girls with anorexia nervosa will be determined.
Detailed description
Adolescence is a critical time for bone mineral accretion as between 60-90% of peak bone mass is established during this period, and peak bone mass is a major determinant of bone density and osteoporosis risk during adulthood. Anorexia nervosa (AN) is the third most common chronic illness among adolescent girls, with a prevalence of 0.2-1.0%. Therefore, AN occurs at a time during which patients are the most vulnerable to disruption of bone mineral accretion. Osteopenia is a major co-morbid complication of AN in 50-75% of female adolescents and adult women with this eating disorder. Women with the onset of the disorder during adolescence have more severe osteopenia than women with onset during adulthood. Little is known about the pathogenesis of osteopenia in this adolescent population and there are no established therapies. Improved understanding of bone mineral metabolism and factors which predict recovery of bone mineral accretion are critical in the development of therapeutic strategies to preserve and maximize bone mass in girls with the onset of AN during adolescence. Estrogen is known to be a critical factor in the development of peak bone mass. Although AN is associated with profound estrogen deficiency, there are no controlled studies investigating the effects of estrogen administration in this population. This research proposal will address critical unanswered questions regarding bone abnormalities in adolescents with anorexia nervosa. Defining changes in bone formation with weight recuperation and hormonal variables would provide insight into the factors essential for bone mineral accretion during adolescence, as well as those that predict recovery. Determination of dose-dependent estrogen effects in this population will be key in targeting interventions during the time of active disease, with the long-term goal of preserving peak bone mass accretion in this vulnerable group of patients. Data obtained from women with post-menopausal osteoporosis or from women with AN cannot be extrapolated to adolescent patients who are in an active period of bone growth and mineralization as well as remodeling. Given the increasing prevalence of AN and its profound consequences on bone health, these studies will provide much needed data to enable treatment strategies for this severe co-morbid disease.
Interventions
DRUGPhysiologic Estrogen/progesterone
Vivelle Dot patch 100 mcg twice weekly; Provera 2.5 mg daily for the first 10 days of the month
OTHERPlacebo
Placebo patches twice weekly; Placebo pills daily for the first 10 days of every month
Sponsors
National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK)
The Hospital for Sick Children
Massachusetts General Hospital
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
QUADRUPLE (Subject, Caregiver, Investigator, Outcomes Assessor)
Eligibility
Sex/Gender
FEMALE
Age
12 Years to 18 Years
Healthy volunteers
Yes
Inclusion criteria
* Females Only with Anorexia Nervosa and Amenorrhea 12-18 years * Normal-weight girls 12-18 years with no past or present history of an eating disorder
Exclusion criteria
* Diseases affecting bone metabolism (including untreated thyroid disease, premature ovarian failure, diabetes, cancer, pituitary, renal disease or bone fracture within the past six months) * Use of prescription medications affecting bone metabolism within three months * Suicidality * Psychosis * Substance abuse * Hematocrit \<30 % * Potassium \<3.0 mmol/L * Glucose \<50 mg/dl.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Percent Change in Spine Bone Density Over the Study Duration (18 Months) | Baseline and 18 months | Bone density at the spine (lumbar 1-4 vertebrae) was measured using dual energy x-ray absorptiometry (DXA) at baseline, 6 months, 12 months and 18 months. The primary outcome was the percent change in bone density at the spine from baseline to 18 months. Areal bone density is measured as g/cm2. The unit of measure for the percent change in bone density is 'percent' Percent change in bone density= \[\[Bone density at 18 months- Bone density at baseline)\*100/Bone density at baseline\]% |
| Change in Spine Bone Mineral Density Z-scores Over the Study Duration (18 Months) | Baseline and 18 months | Bone density at the spine (lumbar 1-4 vertebrae) was measured using dual energy x-ray absorptiometry (DXA) at baseline, 6 months, 12 months and 18 months. The other primary outcome was the change in spine bone density Z-score from baseline to 18 months. The bone density Z-score is a standard deviation score that compares one's bone density to the mean for age and gender, and the Z-score, therefore, does not have any units. It is simply referred to as a Z-score. Change in bone density Z-score= \[Bone density Z-score at 18 months- Bone density Z-score at baseline\] |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Change in N-terminal Propeptide of Type 1 Procollagen (P1NP) Over the Study Duration (18 Months) | Baseline and 18 months | P1NP is a surrogate marker of bone formation that is measured in serum. P1NP levels were measured at baseline, 6, 12 and 18 months. A secondary outcome was the change in P1NP levels from baseline to 18 months: \[P1NP at 18 months - P1NP at baseline). The unit is ng/ml |
Countries
United States
Participant flow
Recruitment details
Period of recruitment: 2003-2009. Location: Massachusetts General Hospital (Boston) and Hospital for Sick Children (Toronto). We screened 150 girls with anorexia nervosa (AN) (110 at MGH and 40 at SickKids) and 88 normal-weight controls 12-18 years. Following the screen, 110 AN and 40 controls were enrolled for the prospective study.
Pre-assignment details
Reasons for enrolled participants being excluded from the trial before assignment to groups primarily included identification of exclusion criteria, loss of interest on the part of the participant, and loss to follow-up.
Participants by arm
| Arm | Count |
|---|---|
| Physiologic Estrogen Replacement Mature girls with anorexia nervosa (bone age 15 or greater): Transdermal estradiol (100 mcg) with cyclic progesterone (days 1-10 of each month).
Immature girls with anorexia nervosa (bone age less than 15 years): Ethinyl estradiol (3.75 mcg daily for the first 6 months, 7.5 mcg daily for the next 6 months, and 11.25 mcg daily for the final 6 months of the study | 55 |
| Placebo Mature girls with anorexia nervosa with a bone age of 15 years of greater: Placebo patches; Immature girls with anorexia nervosa with a bone age of less than 15 years: Placebo pills | 55 |
| Total | 110 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 |
|---|---|---|---|
| Overall Study | Lost to Follow-up | 11 | 12 |
| Overall Study | Physician Decision | 3 | 3 |
| Overall Study | Withdrawal by Subject | 10 | 10 |
Baseline characteristics
| Characteristic | Placebo | Physiologic Estrogen Replacement | Total |
|---|---|---|---|
| Age, Categorical <=18 years | 55 Participants | 55 Participants | 110 Participants |
| Age, Categorical >=65 years | 0 Participants | 0 Participants | 0 Participants |
| Age, Categorical Between 18 and 65 years | 0 Participants | 0 Participants | 0 Participants |
| Age, Continuous | 16.3 years STANDARD_DEVIATION 1.6 | 16.6 years STANDARD_DEVIATION 1.6 | 16.5 years STANDARD_DEVIATION 0.2 |
| Region of Enrollment United States | 55 participants | 55 participants | 110 participants |
| Sex: Female, Male Female | 55 Participants | 55 Participants | 110 Participants |
| Sex: Female, Male Male | 0 Participants | 0 Participants | 0 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | — / — | — / — |
| other Total, other adverse events | 18 / 55 | 20 / 55 |
| serious Total, serious adverse events | 12 / 55 | 16 / 55 |
Outcome results
Primary
Change in Spine Bone Mineral Density Z-scores Over the Study Duration (18 Months)
Bone density at the spine (lumbar 1-4 vertebrae) was measured using dual energy x-ray absorptiometry (DXA) at baseline, 6 months, 12 months and 18 months. The other primary outcome was the change in spine bone density Z-score from baseline to 18 months. The bone density Z-score is a standard deviation score that compares one's bone density to the mean for age and gender, and the Z-score, therefore, does not have any units. It is simply referred to as a Z-score. Change in bone density Z-score= \[Bone density Z-score at 18 months- Bone density Z-score at baseline\]
Time frame: Baseline and 18 months
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Physiologic Estrogen Replacement | Change in Spine Bone Mineral Density Z-scores Over the Study Duration (18 Months) | -0.026 Z -scores | Standard Deviation 0.078 |
| Placebo | Change in Spine Bone Mineral Density Z-scores Over the Study Duration (18 Months) | -0.236 Z -scores | Standard Deviation 0.091 |
Comparison: The null hypothesis was that the groups would not differ for changes in spine bone density z-scores over the study durationp-value: <0.05t-test, 2 sided
Primary
Percent Change in Spine Bone Density Over the Study Duration (18 Months)
Bone density at the spine (lumbar 1-4 vertebrae) was measured using dual energy x-ray absorptiometry (DXA) at baseline, 6 months, 12 months and 18 months. The primary outcome was the percent change in bone density at the spine from baseline to 18 months. Areal bone density is measured as g/cm2. The unit of measure for the percent change in bone density is 'percent' Percent change in bone density= \[\[Bone density at 18 months- Bone density at baseline)\*100/Bone density at baseline\]%
Time frame: Baseline and 18 months
Population: The number of participants was determined using power calculations based on preliminary data. For our primary longitudinal analysis, bone mineral density (BMD) changes were analyzed using a mixed model analysis of variance (intent-to-treat model). For secondary analysis, we examined BMD changes after controlling for age and weight changes.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Physiologic Estrogen Replacement | Percent Change in Spine Bone Density Over the Study Duration (18 Months) | 2.6 Percent change | Standard Deviation 1 |
| Placebo | Percent Change in Spine Bone Density Over the Study Duration (18 Months) | 0.3 Percent change | Standard Deviation 0.1 |
Comparison: Power analysis: We have previously demonstrated that normal female adolescents gain bone density at the rate of 0.039 +/- 0.0507 per year. The pooled SD in that study was 0.046. Based on these data, with a sample size of 110 girls with anorexia nervosa (AN), half of whom are randomized to receive estrogen and half placebo (with a 10% drop-out rate), there will be an 80% chance that we will detect an increase in bone density to 75% of normal in the girls who receive estrogen.p-value: <0.05t-test, 2 sided
Secondary
Change in N-terminal Propeptide of Type 1 Procollagen (P1NP) Over the Study Duration (18 Months)
P1NP is a surrogate marker of bone formation that is measured in serum. P1NP levels were measured at baseline, 6, 12 and 18 months. A secondary outcome was the change in P1NP levels from baseline to 18 months: \[P1NP at 18 months - P1NP at baseline). The unit is ng/ml
Time frame: Baseline and 18 months
Population: The number of participants was determined based on our preliminary data. This analysis was based on completers only.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Physiologic Estrogen Replacement | Change in N-terminal Propeptide of Type 1 Procollagen (P1NP) Over the Study Duration (18 Months) | -2.9 ng/ml | Standard Deviation 22.5 |
| Placebo | Change in N-terminal Propeptide of Type 1 Procollagen (P1NP) Over the Study Duration (18 Months) | -10.9 ng/ml | Standard Deviation 27 |
Comparison: The null hypothesis was that there would be no differences between the groups for changes in P1NP levels over timep-value: >0.05t-test, 2 sided