KEYSTONE, COLO. – Two common comorbidities in type 1 diabetic adolescents are depression and the eating disorder popularly known as diabulimia. Both feature high and upwardly trending hemoglobin A1c levels, often accompanied by weight loss. But a couple of simple questions can help a clinician readily distinguish between the two.
These questions probe for distortions in body image and eating behavior, two core features of bulimia as described in the DSM-IV. These also are central to diabulimia, but diabulimia isn't a DSM-IV diagnosis and thus does not have formal diagnostic criteria, Grace Shih explained at the conference, sponsored by the University of Colorado, Denver, and the Children's Diabetes Foundation at Denver.
A couple of years ago, international experts met in Minneapolis and declared the preferred term for diabulimia is “eating disorder in diabetes mellitus type 1,” abbreviated ED-DMT1. However, “diabulimia” has caught on with patients, families, and the mass media, and Ms. Shih generally stuck with that term in her presentation. Also, because the male:female ratio in diabulimia is 1:10, the same as in patients with DSM-IV eating disorders, Ms. Shih often used “she” in referring to diabulimic teens.
The first question to ask a type 1 diabetic adolescent to learn whether she has a distorted body image is, “How much do you weigh?” The clinician already knows the answer, of course: It's right there in the front of the chart. The question's real purpose is to gain insight into the patient's perception of her weight.
A depressed diabetic teen will often answer, “I don't know my weight, and I don't care.” In contrast, diabulimic patients weigh themselves often. Very often. They may take issue with the accuracy of the clinic scale, according to Ms. Shih, a registered dietician at Packard Children's Hospital at Stanford (Calif.) Medical Center with a practice consisting primarily of type 1 diabetic children and adolescents with eating disorders.
The second question to ask is, “What do you think you should weigh?” The depressed patient may answer, “I don't know – whatever.” But not so with the patient with diabulimia.
“The patients with eating disorders will always say something less than their current weight. They could be thin as a rail, but they see themselves as fat,” she continued.
Diabulimic teens also are somewhat obsessive and compulsive about their eating behavior. They restrict fats and sweets, go on frequent diets, and often feel guilty after eating. And they're exceptionally good at calorie counting. Ask a simple question like, “How many calories are there in an apple?” and the response might be, “How big is the diameter of the apple?”
Depressed diabetic teens, on the other hand, will often eat anything, but in very small quantities. “A bite of chips, a bite of hamburger, a bite of chocolate – that's how they eat,” Ms. Shih said.
Roughly 25% of adolescents with type 1 diabetes have comorbid depression. The prevalence of ED-DMT1 is 11%-30%, depending on the study.
Diabulimia has serious health consequences. Japanese investigators reported that a cohort of type 1 diabetic patients took an average of 11.5 and 15.9 years to develop simple and advanced retinopathy, compared with 3.4 and 7.6 years, respectively, in diabulimic patients. Time to diagnosis of nephropathy averaged 15.1 years in type 1 diabetic patients without diabulimia vs. 6.6 years in those with the eating disorder.
“It's scary, isn't it? But don't try to use this information to scare the teenagers. I've tried. It didn't work,” she said. “Inside a teenager's mind, it's as easy as 1+1=2. Insulin equals weight gain, therefore insulin shots equal fat shots. And no insulin equals weight loss.”
Clinicians with a special interest in ED-DMT1 have adapted the DSM-IV criteria for bulimia, which center on binge eating and “inappropriate compensatory behavior” on at least two occasions per week for 3 months. This compensatory behavior might be purging, misuse of laxatives, or excessive exercise, but in diabulimic patients it takes the form of a greater than 25% reduction in insulin dose for more than 3 months for the purpose of weight loss.
If a patient's blood glucose is in excess of 250 mg/dL, the weight loss can be significant, since the renal threshold for glucose is 160-180 mg/dL.
Ms. Shih cautioned that teens often cheat on their blood glucose records to hide the fact that they're not taking insulin. They may provide recordings of dilute orange juice or dog's blood, but the most common way to cheat is to poke a finger and take a blood sample while running the finger under the faucet.