A previously low BMI was the
most specific warning sign.
Reprinted from Eating Disorders Review
January/February Volume 24, Number 1
©2013 Gürze Books
While anorexia nervosa (AN) patients are known to have a higher-than-normal risk for death, little has been known about mortality among bulimia nervosa (BN) patients. In what is believed to be the largest prospective clinical study of a homogenous sample of adults with severe BN, a group of French researchers led by Dr. Caroline Huas at the University of Paris have identified two indicators of higher risk for death among patients with severe BN (Eur Eat Disord Rev. 2013; 21:15). The study assessed standard mortality rates, causes of death, and predictive factors for death.
The group reviewed articles published between 1980 and 2009 on mortality in the naturalistic outcome of BN patients, and identified only 12 studies had specifically sought to identify causes of mortality among inpatients with BN. These data were then compared against data for the general population. The BN patients had been hospitalized for a number of reasons, including failure of outpatient treatment, serious concurrent general medical problems (such as metabolic abnormalities, hematemesis, and uncontrolled vomiting and life-threatening EKG findings), suicidal tendencies, and severe psychiatric comorbidities. The crude mortality rate was 3.9% for the overall sample of patients, and the mean age at time of death was 29.6 years (range: 19 to 40 years).
Two main predictors of death
When the researchers looked for predictive factors, they identified a history of prior suicide attempts and a lower minimum BMI as significant for a fatal outcome. Suicide was the leading cause of death among the 10 deaths recorded. Two patients died from railway-related suicides, 3 had drug-induced deaths, and one suicide was from an unknown cause. The suicide rate among the BN patients was 6 times greater than that of the general French population.
The other deaths were from lung cancer, and the cause of death of the 3 other patients was unknown. A history of low body mass index (BMI, or kg/m2), was significant, as 9 of the 10 patients had prior histories of AN. None of the 10 patients had children and all were single; for 5 patients, it was the first hospitalization. The authors noted that descriptions of BN patient profiles are scarce because indications for hospitalizations are limited for patients with BN and highly dependent on the health systems of the respective countries. They noted that their study population presented with severe eating disorders because hospitalization in their tertiary care center is a last resort, following clinical guidelines.
The lower previous BMI (a minimum BMI of 17.5 and a DSM-IV diagnosis of AN in the patient’s file) was the most specific risk factor, according to the authors. Noting that the high rate of suicide was similar to that reported for patients with borderline personality disorder, they suggested a future study might describe possible links between BN, personality disorder, and outcome.
