130 Obesity
with diet plus placebo (Davidson et al., 1999). When used
followinga period of low-calorie dieting, orlistat reduces the
regaining of weight lost (Hill et al., 1999).
The major side effects of orlistat include oily spotting, ab-
dominal pain, ”atus with dischar ge, fecal urgency, oily
stools, increased defecation, and fecal incontinence. Side ef-
fects are reported by 20% to 50% of users (Roche Laborato-
ries, 2000). The consumption of excessive quantities of fat
increases the risk of side effects. Thus, in addition to inhibit-
ing fat absorption, the aversive consequences of consuming
fats while taking orlistat may condition patients to limit their
intakes of dietary fats.
The rates of attrition in drug treatment studies have often
been quite high. For example, in the clinical trial of orlistat
by Davidson et al. (1999), more than half the patients in both
the drug (54%) and placebo (57%) conditions dropped out
prior to the “nal evaluation. Moreover, adverse side effects
led to a signi“cantly higher drop out rate among subjects on
orlistat (9%) than on placebo (4%), whereas a lack of treat-
ment effectiveness produced greater attrition among subjects
on placebo (5%) than on orlistat (1%). The combination of a
high attrition rate and differential reasons for subjects drop-
ping out are often not taken into account in analyzing the re-
sults in drug studies. As a consequence, the bene“ts of drug
treatment may be overstated (Williamson, 1999).
An additional concern centers about the use of drugs to
treat obesity independent of signi“cant lifestyle changes.
Many patients, and some practitioners, may rely on medica-
tion as the •magic bulletŽ or sole element of obesity manage-
ment (Kushner, 1997). Such an approach is likely to result in
a disappointing outcome. The bene“ts of weight-loss med-
ications can be enhanced when drug treatment serves as one
component in a comprehensive treatment regimen that in-
cludes lifestyle modi“cation (Wadden, Berkowitz, Sarwer,
Prus-Wisniewski, & Steinberg, 2001).
Bariatric Surgery
Class III or morbid obesity (BMI 40) confers an extremely
high risk for morbidity and decreased longevity. With a
prevalence of 3.9% among women and 1.8% among men,
morbid obesity affects approximately 12 million Americans
(Flegal et al., 1998). Because lifestyle and pharmacological
interventions produce very limited bene“ts for morbidly
obese patients, bariatric surgery represents the treatment of
choice for such individuals (Albrecht & Pories, 1999).
Gastroplasty and gastric bypass are the two major types
of bariatric surgery currently available for morbidly obese
individuals and for persons with BMIs 35 who have
obesity-related comorbid conditions. In vertical banded gas-
troplasty, the stomach is stapled so as create a small vertical
pouch. This gastric pouch limits the amount of food that can be
ingested in a single eating period to about 15 ml. A ring with a
diameter of 9 to 10 mm is placed at the outlet of the pouch to
slow the rate at which food passes through the remainder of the
stomach and into the duodenum and jejunum (small intestine).
Gastroplasty exerts a regulatory effect on eating behavior
through aversive conditioning. Eating more than the small
amount of solid food that the stomach pouch can accommodate
typically results in regurgitation. Fear of vomiting provides a
disincentive for overeating, and the perception of fullness as-
sociated with the distention of the stomach pouch serves as a
cue to stop eating. Unfortunately, gastroplasty does not limit
the consumption of high-calorie liquids or soft foods. As a re-
sult, poor outcome attributable to •soft calorie syndromeŽmay
be as high as 30% (Kral, 1989). An additional problem with
gastroplasty is that over time the size of the pouch may expand,
thereby limiting its long-term effectiveness.
In gastric bypass procedures, such as the Roux-en-Y, a
small gastric pouch is created via stapling, and a limb of the
jejunum is attached directly to the pouch. Ingested food
bypasses 90% of the stomach, the duodenum, and a small
portion of the proximal jejunum (Kral, 1995). The surgery
facilitates weight loss in three ways. First, the pouch can only
hold a small amount of food (15 ml), and over-“lling the
pouch results in regurgitation. Second, the emptying of par-
tially digested food from the pouch into the small intestine re-
sults in malabsorption, such that a portion of nutrients (and
calories) consumed are not absorbed. Third, the consumption
of sweets and foods containing re“ned sugar produces aver-
sive consequences (i.e., the •dumping syndrome) including
nausea, light-headedness, sweating, palpitations, and gas-
trointestinal distress.
Because it produces superior weight-loss outcome, gastric
bypass has replaced gastroplasty as the preferred type of
bariatric surgery (Balsiger, Murr, Poggio, & Sarr, 2000). For
example, Glenny and colleagues (Glenny, O•Meara,
Melville, Sheldon, & Wilson, 1997) reviewed seven studies
that compared gastric bypass with gastroplasty. Six of the
seven showed signi“cantly greater weight losses favoring the
gastric bypass procedure. Typical weight losses one year
after gastric bypass ranged from 45 to 65 kg compared to
30 to 35 kg after gastroplasty. Similar “ndings have been
obtained a large-scale trial of bariatric surgery in Sweden
(C. Sjöstrom, Lissner, Wedel, & Sjöstrom, 1999). Patients
who received gastric bypass had a 33% reduction in body
weight at two years compared to 23% for patients with gas-
troplasty. Long-term studies show some regaining of weight