Medical complications of anorexia nervosa and bulimia nervosa

Medical complications of anorexia nervosa and bulimia nervosa
James E. Mitchella and Scott Crowb
Purpose of review
Introduction
This review focuses on recent publications concerning
Eating disorders are known to result in a variety of
medical complications in patients with eating disorders,
potentially serious medical complications. These are
including anorexia nervosa and bulimia nervosa.
usually most severe in patients with anorexia nervosa
Recent ?ndings
because of the complications attendant to starvation but
Recent literature continues to re?ect that multiple organ
many can also be seen in patients with bulimia nervosa,
systems are frequently affected by eating disorders. The
mainly attributed to the purging behaviors in which these
literature underscores the frequently cited risk of premature
patients engage, including self-induced vomiting and
death in those with anorexia nervosa. A plethora of
laxative abuse [1]. The purpose of this article is to review
dermatologic changes have been described, some
recent literature on the medical complications encoun-
signaling serious underlying pathophysiology, such as
tered in these patients.
purpura, which indicates a bleeding diathesis. Much of the
literature continues to delineate the fact that diabetic
Mortality
patients with eating disorders are at high risk of developing
Although most studies have found substantially increased
diabetic complications. Gastrointestinal complications can
standardized mortality ratios for eating disorders [2], a
be serious, including gastric dilatation and severe liver
recent report by Korndorfer et al. [3] from Olmstead
dysfunction. Acrocyanosis is common, and patients with
County in Minnesota did not ?nd this increase, a ?nding
anorexia nervosa are at risk of various arrhythmias. Low-
that has been of considerable debate in the literature [4].
weight patients are at high risk for osteopenia/osteoporosis.
A recent brief paper by Birmingham et al. [5], however,
Nutritional abnormalities are also common, including
includes a succinct overview of this literature, ?nding that
sodium depletion and hypovolemia, hypophosphatemia and
standardized mortality ratios have varied anywhere from
hypomagnesemia. Resting energy expenditure, although
0.71 to 17.8, and adding data ?nding a standardized mor-
very low in low-weight patients, increases dramatically early
tality ratio of 10.5 (95% con?dence interval ¼ 5.5–15.5),
in refeeding.
again suggesting a markedly exaggerated risk of prema-
Summary
ture death in patients with anorexia nervosa.
Medical complications are common and often serious in
patients with eating disorders, particularly those with
Skin
anorexia nervosa.
An excellent review of the dermatologic signs seen in
patients with eating disorders has recently been published
Keywords
by Strumia [6]. This interesting review documents that a
anorexia nervosa, bulimia nervosa, diabetes, medical
variety of dermatologic abnormalities can be seen in
complications, premature death
patients with eating disorders including xerosis (dry, scaly
skin), lanugo-like body hair (?ne, downy dark hair on the
Curr Opin Psychiatry 19:438–443. ß 2006 Lippincott Williams & Wilkins.
back, abdomen, and forearms), telogen ef?uvium (hair loss
and a positive ‘hair pull test’), acne, carotenoderma
aDepartment of Clinical Neuroscience, University of North Dakota School of
Medicine and the Neuropsychiatric Research Institute, Fargo, North Dakota and
(carotene deposition in the tissues and yellowing of the
bDepartment of Psychiatry, University of Minnesota School of Medicine,
skin because of excess ingestion of carotenoid-rich vege-
Minneapolis, Minnesota, USA
tables), acrocyanosis (circulatory changes resulting in cold,
Correspondence to Professor James E. Mitchell, Neuropsychiatric Research
blue, and occasionally sweaty hands or feet), pruritis,
Institute, 120 South 8th Street, Fargo, ND 58103, USA
Tel: +1 701 365 4916; fax: +1 701 293 3226;
purpura (caused by thrombocytopenia), stomatitis, and
e-mail: mitchell@medicine.nodak.edu
nail dystrophy. Patients with anorexia nervosa and bulimia
Current Opinion in Psychiatry 2006, 19:438–443
nervosa who self-induce vomiting may also demonstrate
Russell’s sign (the presence of scar/callus formation over
Abbreviations
the dorsal surface of the hand, as the hand is used to
ABA
activity-based anorexia
BMDz score
bone mineral density z score
stimulate the gag re?ex to induce vomiting).
BMI
body mass index
FFM
fat-free mass
REE
resting energy expenditure
Endocrine
There has been an ongoing debate in the literature as to
ß 2006 Lippincott Williams & Wilkins
whether eating disorders are more common than would
0951-7367
be expected by chance in girls and young women with
438
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

---

Medical complications of eating disorders Mitchell and Crow
439
type 1 diabetes. Some reports have found higher than
22-year-old borderline mentally retarded man who
expected prevalences and others have not. Mannucci
experienced massive gastric dilatation with evidence of
et al. [7] recently reported a meta-analysis of the con-
mucosal necrosis after binge eating. Lo et al. [14] also
trolled trials on the prevalence of eating disorders in
reported massive gastric dilatation and necrosis in a
patients with type 1 diabetes, which included in total
26-year-old woman with a diagnosis of anorexia nervosa,
748 controls and 1587 women with type 1 diabetes. They
and Barada et al. [15] reported acute gastric dilatation in a
did not ?nd that the prevalence of anorexia nervosa, using
young woman. All three of these cases were managed
either Diagnostic and Statistical Manual of Mental Dis-
conservatively with decompression and did well.
orders, Third Edition, Revised (DSM III-R) or DSM-IV
criteria, was higher among type 1 diabetic patients. They,
Other gastrointestinal problems are frequently seen in
however, did ?nd an increased prevalence of bulimia
these patients. In a study by Winstead and Willard [16],
nervosa and an increased prevalence of the combined
13 patients at an inpatient eating disorders unit were
disorders among type 1 diabetic patients. Treatment
interviewed on their gastrointestinal histories. Eight
issues regarding diabetic patients with eating disorders
(62%) had previously seen a gastroenterologist or primary
were also recently reviewed [8].
care physician for gastrointestinal complaints and of
these, six (46%) had sought treatment for these gastro-
Recently several papers have examined the personality,
intestinal complaints before ever seeking treatment for
temperament, and character in type 1 diabetic patients
their eating disorder. Five (38%) had undergone an
with eating disorders. Pollock-BarZiv and Davis [9]
endoscopy, an upper gastrointestinal barium contrast
studied 51 women with type 1 diabetes intensively using
radiograph, or a lower gastrointestinal barium contrast
interviews and self-report questionnaires. Fourteen of
radiograph. This suggests that family physicians, inter-
the patients appeared to show signi?cant eating disorder
nists, and gastroenterologists should have a high index of
symptoms. Weight preoccupation was associated with the
suspicion about the possible presence of an eating dis-
presence of an eating disorder, and borderline personality
order among young women who present with gastro-
characteristics were related to withholding insulin as
intestinal complaints.
purging mechanism and poor glycemic control. Grylli
et al. [10] screened a sample of 199 adolescents with type
Chial et al. [17] reviewed the manifestations and manage-
1 diabetes for eating disorders, ?nding that those with
ment of anorexia nervosa from the perspective of a
either full syndromal or subthreshhold eating disorders
gastroenterologist. Following an overview of the diagno-
had higher mean scores on a measure of harm avoidance
sis, comorbidity, and multiple organ system dysfunction,
and lower scores on self-directedness. Peveler et al. [11]
the gastrointestinal manifestations were reviewed in
performed a very interesting study in which 87 patients
detail. These included delayed gastric emptying, gastric
who had been interviewed at baseline (age 11–25 years)
motor dysfunction, an impaired sense of hunger and
were recontacted and 63 (72%) were reinterviewed
satiety, delayed small bowel transit time, and consti-
8–12 years later (age 20–38 years) using the Eating
pation, as well as case reports of gastric dilatation, spon-
Disorder Examination. Thirteen individuals met criteria
taneous rupture of the stomach, pancreatitis, necrotizing
at either baseline or follow-up for an eating disorder and
colitis, and perforated ulcer. The paper also provides an
seven evidenced signi?cant eating disorder psycho-
overview of refeeding, and treatment in general.
pathology. Thirty-one (35.6%) misused insulin for weight
control purposes, and there was a strikingly signi?cant
Gendall et al. [18] examined childhood gastrointestinal
relationship between disordered eating habits, insulin
complaints in women with bulimia nervosa. One hundred
misuse, and the risk of developing microvascular com-
and thirty-?ve women with bulimia nervosa were ass-
plications. Grylli et al. [12] also reported that adolescents
essed regarding various gastrointestinal problems. One-
with type 1 diabetes and disordered eating behavior also
third of the participants reported gastrointestinal com-
reported poorer physical and psychosocial quality of life
plaints or constipation in childhood. Women with gastro-
than type 1 diabetic adolescents without disordered
intestinal complaints tended to be younger, to have an ear-
eating. These studies add to a growing body of literature
lier onset of bulimia nervosa, and to have an earlier onset
that suggests that whether or not eating disorders are
of self-induced vomiting, compared with women without
more prevalent in patients with type 1 diabetes, clearly
gastrointestinal complaints. Other interesting recent
the presence of an eating disorder markedly worsens the
reports underscore other possible gastrointestinal prob-
likelihood of an untoward outcome and impaired quality
lems including parotid gland swelling [19], gastritis [20],
of life for patients with type 1 diabetes.
acute liver damage [21], and dental complications [22].
Gastrointestinal complications
Cardiovascular/pulmonary complications
Several case reports of gastric dilatation recently appeared
Cardiovascular and pulmonary complications have been
in the literature. Lunca et al. [13] reported the case of a
well recognized in patients with eating disorders and
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

---

440
Medical comorbidity
include the risk for arrhythmias, acrocyanosis, as noted
?ndings that may have substantial importance for our
above, and pneumomediastinum [1]. Acrocyanosis was
approach to treating anorexia nervosa and its psychiatric
recently studied by Klein-Weigel et al. [23]. Using photo-
comorbidities. In this study, Konstantynowicz et al. [30]
plethysmographic sonography of the brachial artery and
examined bone mineral density in 14 women with
capillary microscopy in symptomatic patients, they de-
anorexia nervosa plus comorbid major depression and
monstrated typical microvascular features of acrocyanosis
31 women with anorexia nervosa but no major depression.
including dilated efferent capillary loops and venoles
BMDz scores were lower (i.e., worse) in individuals with
and reduced capillary ?ow velocities. Roche et al. [24]
anorexia nervosa and comorbid depression as compared
studied abnormalities of the QT interval in patients with
with those without (À2.6 vs. À1.7), and this difference
anorexia nervosa before and after refeeding, ?nding that
was both statistically and clinically signi?cant. In addi-
the QT/RR slope was signi?cantly enhanced in patients
tion, overall level of depression symptoms was inversely
with anorexia nervosa, probably re?ecting autonomic
correlated with total body bone mineral density, as well
imbalance, and that this was reversible after refeeding.
as bone mineral density measured at the lumbar spine.
They stressed the potential clinical importance of this as
These correlations were fairly high (r ¼ À0.4 for total
the enhanced QT/RR slope has also been demonstrated
body and r ¼ À0.6 for lumbar spine).
in patients with life-threatening ventricular arrhythmia.
Krantz and Mehler [25] reported a case of anorexia
Finally, there remain no clearly identi?ed effective
nervosa characterized by tachycardia in which the patient
treatments for diminished bone mineral density in anor-
was found to have indolent lower left extremity cellulites.
exia nervosa, and there is evidence to suggest that weight
As bradycardia is usually seen in these patients, this case
restoration alone is not suf?cient [31]. A recent study
was meant to alert clinicians to the fact that tachycardia
compared alendronate with placebo, given in conjunc-
should prompt a search for other potentially life-
tion with calcium and vitamin D for 1 year [32]. In
endangering conditions. Ohwada et al. [26] reported
this study, alendronate treatment was associated with
evidence of ampulla cardiomyopathy, which is character-
substantial improvements in femoral neck and lumbar
ized by extensive akinesis of the apical region with
spine bone mineral density, whereas placebo treatment
hypercontraction of the basal segment of the ventricle,
was not. As in other studies, low body weight at intake
in three young women with anorexia nervosa, all of whom
into the study predicted lower bone mineral density at
had experienced a hypoglycemic coma. Sundararaghavan
follow-up.
et al. [27] reported an interesting case of the acute onset of
chest pain, which was evaluated in the emergency room
Leptin
and found to be secondary to a spontaneous pneumo-
Leptin is a hormone secreted by fat cells, which appears
mediastinum in a patient with anorexia nervosa.
to have important roles in the regulation of body intake
and body weight. It may be particularly important in food
Skeletal system
deprivation [33] and thus may be important in under-
Several recent studies have examined the issue of the
standing anorexia nervosa.
skeletal complications associated with eating disorders,
particularly anorexia nervosa. These are well recognized
Recent work using a commonly employed animal model
complications [28], but many gaps remain in our knowl-
of anorexia nervosa has helped to expand our under-
edge of the causes, course, and effective treatments for
standing of the way in which leptin might work in
these. A recent report by Misra et al. [29] reinforces the
individuals with anorexia nervosa [34]. This study
high frequency with which such complications occur.
examined rats with activity-based anorexia (ABA) that
These investigators examined 60 individuals presenting
developed a combination of physical hyperactivity plus
with anorexia nervosa and compared them with 58 normal
food refusal. Previous work has suggested that leptin
controls. Bone mineral density was measured and the
administration reduces the degree of semistarvation-
results show z scores less than or equal to À1 in 41% of the
induced hyperactivity seen in this animal model [35].
individuals with anorexia nervosa, and less than or equal
This study compared rats with ABA with those allowed to
to À2 in 11% of the anorexia nervosa sample. This was in
exercise ad-lib and with rats that were sedentary. All
contrast with the control sample in which bone mineral
three groups received intracerebral ventricular leptin. In
density z (BMDz) scores less than or equal to À1 were
ABA rats, decreased running-wheel activity, diminished
found in 23% (for BMDz < À2, 2.0%). In this study, body
food intake, and increased temperature were seen. No
mass index (BMI), amount of lean body mass, and age at
changes in activity were seen in the ad-lib or sedentary
menarche were all predictors of diminished bone mineral
groups, with only minimal changes in temperature, but
density.
food intake was decreased. This complex set of ?ndings
(diminished hyperactivity in ABA rats but also dimin-
Another study also recently examined the question of
ished food intake) may serve to temper enthusiasm about
predictors of bone mineral density, with the interesting
leptin as a treatment for anorexia nervosa.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

---

Medical complications of eating disorders Mitchell and Crow
441
Another study has attempted to examine the question of
estimating changes in body composition and energy
leptin function in humans with anorexia nervosa [36].
expenditure, understanding marked changes in weight
This study examined 23 women with anorexia nervosa
that can occur in treatment, and how to best monitor the
and 21 controls, all adolescents, and involved frequent
changes in body compartments. Four recent papers have
sampling of leptin, growth hormone, cortisol, and ghrelin.
helped to illuminate further these areas. In the ?rst of
Detailed studies of leptin release showed lower amounts
these, Van Wymelbeke et al. [40] followed resting energy
of pulsatile and total leptin release in anorexia nervosa,
expenditure (REE) at baseline and three points over the
and ?ne-grained analysis showed that this was because of
?rst 7 weeks of treatment in 87 women with anorexia
lower levels of basal and burst leptin release.
nervosa. These investigators found that REE was sub-
stantially increased by day 8; the amount of this increase
Nutritional issues
was substantially out of proportion to the change in fat-
Historically, substantial attention has been paid to the nu-
free mass (FFM) observed (13.4% change in REE vs.
tritional aspects of anorexia nervosa and its treatment, and
1.8% change in FFM). Throughout the rest of the treat-
three recent reports helped to expand our knowledge in
ment, this marked disparity between changes in REE and
this area. First, Caregaro et al. [37] published a recent
FFM persisted.
report highlighting the importance of sodium depletion
and hypovolemia and associated hemoconcentration in
Regional body composition and hormonal function are
anorexia nervosa. They reported a sample of 14 individuals
clearly linked, but the nature of this complex relationship
hospitalized for a mean of 24 days for anorexia nervosa
has not been well understood. Misra et al. [41] examined a
treatment. Mean BMI at admission was 12.9, and at dis-
sample of 23 individuals with anorexia and 20 normal
charge, 14.2. Sixty-four percent of the sample had hypo-
controls and attempted to correlate hormonal function
volemia at intake laboratory assessment. In spite of this
with changes in body composition. Not surprisingly,
(or perhaps more accurately, because of this), hematolo-
differences were seen between controls and individuals
gic measures, such as hematocrit, hemoglobin among
with anorexia nervosa in terms of percentage of trunk fat,
others, were generally within the normal range at admis-
trunk/extremity fat ratio, trunk lean mass, and trunk/
sion, but with adequate hydration, anemia became appar-
extremity lean-mass ratio. Interestingly, in the normal
ent in six out of the 14 individuals in this sample. The
control group the area under the curve for growth hor-
authors noted that such hemoconcentration due to hypo-
mone was inversely correlated with trunk fat and trunk
volemia may lead to falsely reassuring laboratory values
extremity fat ration, but no such relationship was seen in
when individuals present for eating disorders treatment.
anorexia nervosa. Nadir levels of cortisol, however, were
inversely associated with extremity lean mass and
Electrolyte disturbances have received attention in this
directly correlated with trunk lean mass. The magnitude
area for long, but in the treatment of anorexia nervosa,
of these correlations was substantial (for external lean
hypophosphatemia has received the greatest emphasis.
mass, r ¼ 0.49, and for trunkal lean mass, r ¼ 0.48).
Birmingham et al. [38] reported a series of 50 individ-
uals admitted for anorexia nervosa treatment; of these 50,
For many patients the pace of weight gain in treatment is
30 (60%) developed hypomagnesemia at some point
steady, but on occasions marked changes in weight can be
during their hospitalization. Of note, only 16% of them
seen. Yucel et al. [42] published a series of two cases in
had it at admission, whereas the others developed it
which marked weight gain (in one case 9 kg, in the other
during treatment, as late as 3 weeks into treatment.
case 3 kg) was seen in the ?rst week. These investigators
These results suggest the importance of continued moni-
emphasized the importance of this kind of edematous
toring of magnesium throughout the early and middle
weight gain for clinical treatment.
stages of treatment in order to detect the possible emer-
gence of later in treatment hypomagnesemia.
Finally, as weight changes, there is some interest, both
clinical and in research, in measuring the size of different
Finally, a case report serves to emphasize the potential
body compartments. Piccoli et al. [43] compared bio-
severity of nutritional de?ciencies that can be encoun-
electrical-impedance analysis with skin-fold thickness
tered. This report described a 33-year-old woman with an
measurements in 74 women with anorexia nervosa. This
eating disorder who had developed as a result of that
study showed poor agreement between these methods
vitamin A de?ciency leading to corneal disease culminat-
in individuals below a BMI of 15, and these authors
ing in blindness [39].
cautioned strongly against the use of such measures in
these low-weight individuals.
Metabolism and body composition
A number of important clinical issues are encountered
Conclusion
in the area of weight, energy expenditure, and refeed-
Anorexia nervosa and bulimia nervosa represent psy-
ing. These include problems with understanding and
chiatric disorders wherein medical complications are
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

---

442
Medical comorbidity
common, and in the case of anorexia nervosa, expected.
15 Barada KA, Azar CR, Al-Kutoubi AO, et al. Massive gastric dilatation after a
single binge in an anorectic woman. Int J Eat Disord 2006; 39:166–169.
In patients with anorexia nervosa every major organ
16 Winstead NS, Willard SG. Frequency of physician visits for GI complaints by
system is involved, and the risk of mortality is substantial.
anorexic and bulimic patients. Am J Gastroenterol 2001; 96:1667–1668.
Particular areas of concern are highlighted in the article
17 Chial HJ, McAlpine DE, Camilleri M. Anorexia nervosa: manifestations and
and include dermatologic changes (some of which evi-
management for the gastroenterologist. Am J Gastroenterol 2002; 97:255–
269.
dently need acute intervention; e.g., purpura), endocrine
18 Gendall KA, Joyce PR, Carter FA, et al. Childhood gastrointestinal complaints
abnormalities (including mismanagement of diabetes),

in women with bulimia nervosa. Int J Eat Disord 2005; 37:256–260.
gastrointestinal problems (including the risk of gastric dil-
Interesting study documenting increased gastrointestinal complaints during child-
hood in a subgroup of women with eating disorders.
atation), cardiovascular/pulmonary problems (including
19 Mandel L, Abai S. Diagnosing bulimia nervosa with parotid gland swelling.
arrhythmias and pneumomediastinum), severe electro-
J Am Dent Assoc 2004; 135:613 –616 [quiz 655].
lyte abnormalities, and bone demineralization. Physi-
20 Sansone RA, Naqvi A, Sansone LA. An unusual cause of dizziness in bulimia
nervosa: a case report. Int J Eat Disord 2005; 37:364–366.
cians caring for these patients must be aggressive in
21 Di Pascoli L, Lion A, Milazzo D, Caregaro L. Acute liver damage in anorexia
pursuing, and when they are found in treating, these
nervosa. Int J Eat Disord 2004; 36:114–117.
potentially life-endangering complications.
22 Frydrych AM, Davies GR, McDermott BM. Eating disorders and oral health:
a review of the literature. Aust Dent J 2005; 50:6 –15 [quiz 56].
References and recommended reading
23 Klein-Weigel P, Rein P, Kronenberg F, et al. Microcirculatory assessment of
Papers of particular interest, published within the annual period of review, have
vascular acrosyndrome in anorexia nervosa and analysis of manifestation
been highlighted as:
factors. J Psychosom Res 2004; 56:145–148.

of special interest
24 Roche F, Barthelemy JC, Mayaud N, et al. Refeeding normalizes the QT rate

of outstanding interest

dependence of female anorexic patients. Am J Cardiol 2005; 95:277–280.
Additional references related to this topic can also be found in the Current
Study examining QT interval changes in patients with anorexia nervosa.
World Literature section in this issue (pp. 460–461).
25 Krantz MJ, Mehler PS. Resting tachycardia, a warning sign in anorexia
nervosa: case report. BMC Cardiovasc Disord 2004; 4:10.
1
Birmingham CL, Beumont P. Medical management of eating disorders.
New York: Cambridge University Press; 2004.
26 Ohwada R, Hotta M, Kimura H, et al. Ampulla cardiomyopathy after hypogly-
cemia in three young female patients with anorexia nervosa. Intern Med 2005;
2
Sullivan PF. Mortality in anorexia nervosa. Am J Psychiatry 1995; 152:1073 –
44:228–233.
1074.
27 Sundararaghavan S, Pitts TY, Suarez WA, Johnstone C. Chest pain among
3
Korndorfer SR, Lucas AR, Suman VJ, et al. Long-term survival of patients with
adolescents with anorexia nervosa. Pediatr Emerg Care 2005; 21:603–605.
anorexia nervosa: a population-based study in Rochester. Minn Mayo Clin
Proc 2003; 78:278–284.
28 Herzog W, Minne H, Deter C, et al. Outcome of bone mineral density in
anorexia nervosa patients 11.7 years after ?rst admission. J Bone Miner Res
4
Sullivan PF. Discrepant results regarding long-term survival of patients with
1993; 8:597–605.
anorexia nervosa? Mayo Clin Proc 2003; 78:273–274.
29 Misra M, Aggarwal A, Miller KK, et al. Effects of anorexia nervosa on clinical,
5
Birmingham CL, Su J, Hlynsky JA, et al. The mortality rate from anorexia
hematologic, biochemical, and bone density parameters in community-dwell-

nervosa. Int J Eat Disord 2005; 38:143–146.
ing adolescent girls. Pediatrics 2004; 114:1574–1583.
Fairly brief report that nonetheless manages to provide an interesting review of the
literature on mortality issues in anorexia nervosa and also includes a new dataset
30 Konstantynowicz J, Kadziela-Olech H, Kaczmarski M, et al. Depression in
suggesting an increased mortality rate.

anorexia nervosa: a risk factor for osteoporosis. J Clin Endocrinol Metab 2005;
90:5382 –5385.
6
Strumia R. Dermatologic signs in patients with eating disorders. Am J Clin
A paper describing increased rates of osteoporosis when depression and anorexia

Dermatol 2005; 6:165–173.
nervosa occur together.
Excellent, although technical, review of dermatologic abnormalities seen in patients
with eating disorders.
31 Bachrach LK, Katzman DK, Litt IF, et al. Recovery from osteopenia in
adolescent girls with anorexia nervosa. J Clin Endocrinol Metab 1991;
7
Mannucci E, Rotella F, Ricca V, et al. Eating disorders in patients with type 1
72:602–606.

diabetes: a meta-analysis. J Endocrinol Invest 2005; 28:417–419.
Meta-analysis suggesting no increase in the prevalence of eating disorders among
32 Golden NH, Iglesias EA, Jacobson MS, et al. Alendronate for the treatment of
patients with type 1 diabetes mellitus.

osteopenia in anorexia nervosa: a randomized, double-blind, placebo-con-
trolled trial. J Clin Endocrinol Metab 2005; 90:3179 –3185.
8
Goebel-Fabbri AE, Fikkan J, Connell A, et al. Identi?cation and treatment of
A well designed osteoporosis treatment trial.
eating disorders in women with type 1 diabetes mellitus. Treat Endocrinol
2002; 1:155–162.
33 Chan JL, Mantzoros CS. Role of leptin in energy-deprivation states: normal

human physiology and clinical implications for hypothalamic amenorrhoea and
9
Pollock-BarZiv S, Davis C. Personality factors and disordered eating in
anorexia nervosa. Lancet 2005; 366:74–85.
young women with type 1 diabetes mellitus. Psychosomatics 2005; 46:
A useful review of leptin in anorexia nervosa.
11–18.
34 Hillebrand JJ, Koeners MP, de Rijke CE, et al. Leptin treatment in activity-
10 Grylli V, Hafferl-Gattermayer A, Wagner G, et al. Eating disorders and eating
based anorexia. Biol Psychiatry 2005; 58:165–171.
problems among adolescents with type 1 diabetes: exploring relationships
with temperament and character. J Pediatr Psychol 2005; 30:197–206.
35 Exner C, Hebebrand J, Remschmidt H, et al. Leptin suppresses semi-starva-
tion induced hyperactivity in rats: implications for anorexia nervosa. Mol
11 Peveler RC, Bryden KS, Neil HA, et al. The relationship of disordered eating
Psychiatry 2000; 5:476–481.

habits and attitudes to clinical outcomes in young adult females with type 1
diabetes. Diabetes Care 2005; 28:84–88.
36 Misra M, Miller KK, Kuo K, et al. Secretory dynamics of leptin in adolescent
Very interesting study, one of the few including longitudinal data, on patients with
girls with anorexia nervosa and healthy adolescents. Am J Physiol Endocrinol
disordered eating and diabetes mellitus.
Metab 2005; 289:E373–E381.
12 Grylli V, Wagner G, Hafferl-Gattermayer A, et al. Disturbed eating attitudes,
37 Caregaro L, Di Pascoli L, Favaro A, et al. Sodium depletion and hemocon-

coping styles, and subjective quality of life in adolescents with type 1 diabetes.

centration: overlooked complications in patients with anorexia nervosa?
J Psychosom Res 2005; 59:65–72.
Nutrition 2005; 21:438–445.
One of the few studies that carefully examine quality of life in patients with eating
A study that demonstrates the role of dehydration in falsely ‘normalizing’ laboratory
disorders and diabetes mellitus.
parameters in anorexia nervosa.
13 Lunca S, Rikkers A, Stanescu A. Acute massive gastric dilatation: severe
38 Birmingham CL, Puddicombe D, Hlynsky J. Hypomagnesemia during refeed-
ischemia and gastric necrosis without perforation. Rom J Gastroenterol 2005;
ing in anorexia nervosa. Eat Weight Disord 2004; 9:236–237.
14:279–283.
39 Velasco Cruz AA, Attie-Castro FA, Fernandes SL, et al. Adult blindness
14 Lo DY, Yen JL, Jones MP. Massive gastric dilation and necrosis in anorexia
secondary to vitamin A de?ciency associated with an eating disorder. Nutri-
nervosa: cause or effect? Nutr Clin Pract 2004; 19:409–412.
tion 2005; 21:630–633.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

---

Medical complications of eating disorders Mitchell and Crow
443
40 Van Wymelbeke V, Brondel L, Marcel Brun J, Rigaud D. Factors asso-
42 Yucel B, Ozbey N, Polat A, Yager J. Weight ?uctuations during early refeeding
ciated with the increase in resting energy expenditure during refeeding in
period in anorexia nervosa: case reports. Int J Eat Disord 2005; 37:175–177.
malnourished anorexia nervosa patients. Am J Clin Nutr 2004; 80:1469 –
43 Piccoli A, Codognotto M, Di Pascoli L, et al. Body mass index and agreement
1477.

between bioimpedance and anthropometry estimates of body compartments
41 Misra M, Miller KK, Almazan C, et al. Hormonal determinants of regional body
in anorexia nervosa. JPEN J Parenter Enteral Nutr 2005; 29:148–156.
composition in adolescent girls with anorexia nervosa and controls. J Clin
This study demonstrates the challenges in assessing altered body composition in
Endocrinol Metab 2005; 90:2580 –2587.
individuals with anorexia nervosa or low weight.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

---

Document Outline

• Medical complications of anorexia nervosa and bulimia nervosa
  • Introduction
  • Mortality
  • Skin
  • Endocrine
  • Gastrointestinal?complications
  • Cardiovascular/pulmonary?complications
  • Skeletal?system
  • Leptin
  • Nutritional?issues
  • Metabolism and body?composition
  • Conclusion
  • References and recommended reading

---