by Timothy D. Brewerton, MD, DFAPA, FAED
It is well known that the optimal treatment of eating disorders, especially anorexia nervosa (AN), often requires a unified team approach consisting of patients, family members, and multiple health care providers, including primary care doctors. These primary care doctors may be trained in family medicine, pediatrics, internal medicine, or obstetrics and gynecology, and their training, knowledge and experience with eating disorders often varies considerably. As a psychiatrist who specializes in eating disorders in children, adolescents and adults, I have worked with all kinds of doctors and a variety of other health care professionals as well. One of the most challenging issues for me when working with patients and their other doctors is when birth control pills (BCPs) or oral contraceptives are prescribed for the treatment of amenorrhea (absence of menstrual periods) or oligomenorrhea (infrequent or irregular periods) accompanying an eating disorder. In a minority of cases, the BCPs were prescribed prior to the onset of menstrual irregularities, and were simply not discontinued after eating disorder onset. But in most cases the BCPs are prescribed during the active course of the illness. Although these physicians mean well for their patients, they are making a very big mistake that can compromise full recovery. Firstly, they are removing from consideration the one best indicator of normal weight attainment for adolescent girls and premenopausal women, and secondly, there is no evidence whatsoever that BCPs help or prevent osteoporosis or osteopenia (which is the primary reason they are prescribed). In fact, there is lots of evidence that BCPs don’t work for this common complication. In fact, five separate trials have shown estrogen therapy to be of no benefit for decreased bone mineral density in AN patients (Mehler & MacKenzie, 2009).
Although every physician is taught to treat the underlying cause of presenting problems, or symptoms, the cause is not always readily identified, evident or known. It is no secret that nonpsychiatric physicians often miss or misdiagnose any number of psychiatric disorders, including eating disorders. Even when it seems evident that a patient is severely underweight and has AN, some doctors are hesitant to “call a spade a spade.” This is because they don’t understand the disorder or they are uncomfortable making a psychiatric diagnosis for fear of offending the patient. Some doctors believe the fallacy that psychiatric disorders are disorders of exclusion and that every medical disorder known to man must be ruled out first. As a result the patient and family are sometimes put through unnecessary, uncomfortable, risky and expensive procedures, e.g., upper gastrointestinal endoscopy, MRI of the brain, etc., to rule out unlikely disorders. Psychiatric disorders, including eating disorders, all have very specific criteria as defined in the DSM-IV that need to be met. They are all diagnoses of inclusion; one just needs to know what questions to ask (both patients and family members) in order to demonstrate the presence of the disorder.
Once the proper diagnosis or diagnoses are established then evidence-based treatment can commence. In the case of AN, resumption of eating (refeeding) and restoration of normal weight is essential for recovery. Return of menstrual function is a necessary prerequisite for the resolution of the underlying psychological symptoms of eating disorders, such as disturbed body perception and obsession with shape and weight (Clausen et al., 2004). Usually there is a goal weight or goal weight range that is set by the leader of the treatment team in an attempt to reverse the effects of starvation. There are a number of ways to calculate goal weight, depending on age and gender, but these include the use of CDC growth charts (weight, height and BMI for age and gender) and the use of Metropolitan Life Insurance average body weight and height charts (1959 and 1983 versions) (Metropolitan Life Insurance Company, 1959; 1983). Bryan Lask and colleagues have proposed the use of ovarian ultrasound to identify ovarian follicle formation as a measure of follicular maturation and imminent menses (Allan et al., 2010). Whatever method is used, the final common goal is resumption of menses in women and girls given that in premenopausal women and post-menarchal (after the first menses) girls the most reliable sign that weight restoration has been successfully attained is when menstrual periods resume and continue on time (every 24-35 days) for at least 2-3 months.
The weight necessary to resume menses can be estimated with great accuracy and success by using the Frisch tables, which are surprisingly not well known to many physicians (Frisch et al., 1974; Frisch, 2004). There are two Frisch tables, one for girls with primary amenorrhea (never had a period) and one for girls with secondary amenorrhea (began then lost menses). The second table (secondary amenorrhea) is the one that should be used for ED patients, including those who have never had a period before (pre-menarchal). Usually I will use the 50th percentile weight for height in determining goal weight range for any given individual with AN, which indicates a 50-50 chance of menstrual function returning at this weight for height. If menses have not returned after maintaining this weight for 3-6 months, then I recommend increasing the goal weight range by 5 lbs, and again wait for 3-6 months for results. Conversely, if menses commence at a weight lower than goal weight, then the goal weight range can be adjusted downward.
As a psychiatrist specializing in the treatment of eating disorders for most of my career, I can say that this approach works in at least 9 out of 10 times. Very commonly, patients do not want to stop their BCPs and risk having to gain more weight. Instead they prefer the false sense of security that having monthly blood flow brings them, their families and their uninformed health care providers. In summary, 1) Do not allow birth control pills (BCPs) to be prescribed to young women or girls with AN until after full restoration of weight and menses occurs (at least 2-3 cycles on time); 2) BCPs obscure the most reliable indicator of healthy weight recovery for adolescent girls and pre-menopausal women, i.e., restoration of menstrual function; 3) BCPs do absolutely nothing to prevent osteoporosis or osteopenia, which is driven by many other factors related to starvation besides low estrogen levels, e.g., alterations in cortisol, insulin-like growth factor, and leptin (Mehler & MacKenzie, 2009)
Submitted as an article for the F.E.A.S.T. Online Newsletter September 27, 2010
Dr. Timothy D. Brewerton is a Clinical Professor of Psychiatry and Behavioral Sciences at the Medical University of South Carolina, Charleston.