Hypercortisolism and Its Presence in Patients With Eating Disorders

Hypercortisolism (or Cushing’s syndrome) is sometimes present in those with an eating disorder. This state can occur either during active restriction or in the early phases of re-feeding. Cushing’s disease specifically involves excess cortisol in the body due to a pituitary tumor, all other types of hypercortisolism are classified as hypercortisolism or Cushing’s syndrome. Cushing’s disease is rare (1.2 to 2.4 cases per million annually).

Flickr.com: Briend

Flickr.com: Briend


No one really knows the exact mechanism for this hypothalamic-pituitary-adrenal axis abnormality that shows up in those with eating disorders, but it is presumed to be helpful in the conservation of energy. In other words, it likely keeps you alive in the short term.

Two things to know about a possible diagnosis of hypercortisolism if you are dealing with an active eating disorder and/or early in a recovery process:

  1. It’s much more common in those who are active and exercising, so if you are still exercising stop. 1
  2. It will commonly resolve with continued rest and re-feeding. 2

What this means is that you and your doctor will want to discuss exactly what the levels suggest right now and whether it is possible to take a watchful waiting approach for the next 3-6 months while you make sure you are eating fully to your hunger and not exercising or working out at all.

A serum cortisol level test tells us how much cortisol is circulating in your blood. The problem with serum cortisol level tests is that having high circulating levels in your blood could be the result of being under stress or your body fighting off infection at the time the blood was taken. The normal range for the serum cortisol level test is 6 to 23 mcg/dL, but that varies between labs.

When a serum cortisol test is outside the normal range, a physician will likely follow up with a requisition for you to undergo a urine test and the dexamethasone suppression test. And that is because a definitive diagnosis of hypercortisolism is not based on the serum cortisol test alone. However, before you just go ahead with further workup to rule in or out hypercortisolism (Cushing’s syndrome), have a frank discussion with your doctor about the status of your eating disorder.

Should you have an elevated serum cortisol level and subsequent screening confirms hypercortisolism, then you might feel rising panic. And that panic will, in large part, be associated with the socioculturally-driven terror induced by the correlation of hypercortisolism and obesity— it’s emphasized all over mainstream medical online sites. And the problem with all things obesity research, is the absolute blithe disregard for the difference between correlation and causation.

The number of papers that begin with the fallacy-as-statement: “obesity has reached epidemic levels,” is significant. Google Scholar returns approximately 2700 peer-reviewed published papers using that exact unscientific phrase. And it’s a fallacy because an epidemic in this non-medical context means “extremely prevalent and rapidly increasing.” Obesity rates have not been increasing, let alone rapidly increasing and the prevalence remains that 2/3 of our population is not of above-average weight at all. 3 In fact, if anyone is quoting data regarding an increase in obesity rates, it will be seriously out of date.

So much has the sociocultural hatred of fat permeated scientific literature, it requires sifting through shrill unscientific introductions, discussions and conclusions to isolate fact from fiction. I’ll use the discussion in one systematic review to illustrate this point:

“In the increasingly widespread calorie-dense environments that most humans inhabit, it is uncertain whether HPA [hypothalamic-pituitary-axis] dysregulation is driving obesity or whether obesity perturbs the HPA axis…Given substantial evidence of hypercortisolism in obesity and evidence that increased cortisol reactivity among those with abdominal obesity (not general obesity) may endure even for days (Epel. et al., 2000), we suspect that it is possible that increased cortisol concentrations may at least perpetuate abdominal obesity, if not cause it." 4

Would you like to know what that “substantial evidence” from Epel and colleagues might be? It involved 59 healthy premenopausal women, 30 with a high waist-to-hip ratio and 29 with a low waist-to-hip ratio. Or, if you’d prefer, 30 women who were decidedly apple-shaped and 29 who were not. They underwent various stress sessions and the apple-shaped women found these sessions more threatening, performed more poorly on them and reported more chronic stress than their 29 low-waist-to-hip-ratio counterparts. The “apples” also secreted significantly more cortisol during the first stress session than their counterparts (that means that despite their subjective experience, after the first session they were not secreting more cortisol than their counterparts). It was only the thin women with a high-waist-to-hip-ratio who continued to secrete significantly more cortisol to the challenges through repeated exposures. 5

Guess what that one small study actually appears to indicate? It suggests that women with high-waist-to-hip-ratios have more sensitive initial responses to stressors than their low-waist-to-hip ratio counterparts, but that only those who have a high-waist-to-hip-ratio who are also thin actually have persistent high cortisol levels in the presence of stress to which they are habituated. It is most decidedly not “substantial evidence” that increased cortisol reactivity among those with abdominal fat endures for days, given that the above-average weight “apples” (i.e. those who were classified as obese) had no persistent cortisol response, merely an initial spike. As for abdominal fat perpetuating cortisol concentrations? No. And causation? Oh, puhlease.

If anything, this small study (and I cannot emphasize enough how far it is from offering anything definitive) suggests that hypercortisolism is most persistent for those who are thin “apples” under stress. That certainly dovetails more solidly with the fact that starvation (dieting) is stressful and hypercortisolism is an adaptive response to stress that is unrelenting and yet not something to which we can habituate. And yes, there are naturally thin apple-shaped women in the world but their prevalence will be less than 20% of the 2-4% of the overall population who are naturally thin. In other words, it is far more likely that the thin apple-shaped women in the study were active dieters than it is likely they were naturally thin.

Just because we can measure it, doesn’t make it a disease. The body is a living ecosystem responding to all the vagaries of internal and external stimuli. In the absence of a pituitary tumor (Cushing’s disease) or an adrenocortical adenoma (a tumor impacting the adrenal glands), the most likely cause of hypercortisolism is unrelenting stress. Framing an adaptive response to stress as a disease state overlooks the complexities of why hypercortisolism might be present in any individual in the first place.

The systematic reviews looking at the treatment of hypercortisolism with various prescriptions or an adrenalectomy aren’t offering up promising results either. 6, 7 When incidental benign adrenocortical adenomas are identified in conjunction with hypercortisolism then the greater the post dexamethasone cortisol level, the greater the hazard mortality ratio. The hazard mortality ratio was 12 for cortisol levels of <1.8 µg/dL and 22 for cortisol levels of >5µg/dL. But actual mortality for this group overall was 6% of those in the study. Death was attributed to complications of cardiovascular disease and infection. 8 In other words, even when hypercortisolism appears correlated to the presence of a benign tumor, it still correlates most strongly to remaining alive, rather than dying from presumed complications.

Does lowering the body’s cortisol levels remove any environmental stress? Obviously not. Does lowering serum cortisol levels restore energy balance for a patient that is chronically increasing energy deficits through either/or under eating and over exertion relative to food intake levels? No, again.

All this brings us back to the main topic at hand: an eating disorder. If you are actively restricting intake, or generating progressive energy deficits through a combination of under eating and excess exertion and activity, then if you are diagnosed with hypercortisolism you and your doctor need to have a frank discussion of risks and benefits associated with treating hypercortisolism, if recovery is not underway for you at this time. Keeping in mind that the medical treatment options available are not promising, it may be more worthwhile to at least remove exercise from your restrictive regime as a way to ease the symptoms associated with hypercortisolism.

If you are undertaking a recovery effort, then it’s best to ask your doctor the following question: “So what if we do nothing for another 6 months? If I choose to wait and re-test at that time are you comfortable with that approach or are there risks in going that approach that you want me to know about before I make my decision?”

As mentioned at the outset of this paper, most hypercortisolism that appears in recovery will resolve of its own accord with continued re-feeding and complete rest. It’s helpful to evaluate environmental stressors in your life and do what you can to remove the stressors that can be removed. However, it’s also important to accept that not all stressors are modifiable in life. Be compassionate towards yourself and don’t assume hypercortisolism reflects a body that is failing. We actually don’t know how much quicker people with unrelenting stress would die were there no hypothalamic-pituitary axis response in the first place. We get our correlations wrong all the time and we are most certainly nowhere near understanding causation when it comes to how any individual navigates stress on a biological level.


1. Klein, Diane Alix, Laurel ES Mayer, Janet Ellen Schebendach, and B. Timothy Walsh. "Physical activity and cortisol in anorexia nervosa." Psychoneuroendocrinology 32, no. 5 (2007): 539-547.

2. Winterer, Jorg, Happy E. Gwirtsman, David T. George, Walter H. Kaye, D. Lynn Loriaux, and Gordon B. Cutler Jr. "Adrenocorticotropin-stimulated adrenal androgen secretion in anorexia nervosa: impaired secretion at low weight with normalization after long-term weight recovery." The Journal of Clinical Endocrinology & Metabolism 61, no. 4 (1985): 693-697.

3. Flegal, Katherine M., Margaret D. Carroll, Brian K. Kit, and Cynthia L. Ogden. "Prevalence of obesity and trends in the distribution of body mass index among US adults, 1999-2010." Jama 307, no. 5 (2012): 491-497.

4. Incollingo, Rodriguez AC, E. S. Epel, M. L. White, E. C. Standen, J. R. Seckl, and A. J. Tomiyama. "Hypothalamic-pituitary-adrenal axis dysregulation and cortisol activity in obesity: A systematic review." Psychoneuroendocrinology 62 (2015): 301-318.

5. Epel, Elissa S., Bruce McEwen, Teresa Seeman, Karen Matthews, Grace Castellazzo, Kelly D. Brownell, Jennifer Bell, and Jeannette R. Ickovics. "Stress and body shape: stress-induced cortisol secretion is consistently greater among women with central fat." Psychosomatic medicine 62, no. 5 (2000): 623-632.

6. Gadelha, Mônica R., and Leonardo Vieira Neto. "Efficacy of medical treatment in Cushing's disease: a systematic review." Clinical endocrinology 80, no. 1 (2014): 1-12.

7. Graversen, D., Vestergaard, P., Stochholm, K., Gravholt, C.H. and Jørgensen, J.O.L., 2012. Mortality in Cushing's syndrome: a systematic review and meta-analysis. European journal of internal medicine, 23(3), pp.278-282.

8. Debono, Miguel, Mike Bradburn, Matthew Bull, Barney Harrison, Richard J. Ross, and John Newell-price. "Cortisol as a marker for increased mortality in patients with incidental adrenocortical adenomas." The Journal of clinical endocrinology and metabolism 99, no. 12 (2014): 4462-4470.