Obesity Surgery, 11, 491-495
Outcome of Biliopancreatic Diversion in Subjects
with Prader-Willi Syndrome
Giuseppe M. Marinari; Giovanni Camerini; Giorgio Baschieri Novelli;
Francesco Papadia; Federica Murelli; Paola Marini; Gian Franco
Adami; Nicola Scopinaro
Semeiotica Chirurgica R, University of Genoa School of Medicine, Genoa, Italy
Background: In Prader-Willi Syndrome (PWS), mental retardation and compulsive hyperphagia cause
early obesity, the co-morbidities of which lead to
short life-expectancy, with death usually occurring in
their 20s. Long-term weight loss is mandatory to
lengthen the survival; therefore, the lack of compliance in voluntary food restriction requires a surgical
malabsorptive approach.
Methods: 15 PWS subjects were submitted to biliopancreatic diversion (BPD) and followed (100%) for
a mean period of 8.5 (4-13) years. BPD consists of a
distal gastrectomy with a long Roux-en-Y reconstruction which, by delaying the meeting between
food and biliopancreatic juices, causes an intestinal
malabsorption. Indication for BPD was BMI >40 or
>35 with metabolic complications. Preoperative
mean age was 21±5 years, mean weight 127±26 kg,
and mean Body Mass Index (BMI, kg/m2) 53±10.
According to Holm’s criteria, all of the subjects had
a total score ³8. IQ assessment was performed in
each subject, with a mean score of 72±10. An arbitrary lifestyle score was given to each subject.
Results: No perioperative complications were
observed. Percent excess weight loss (%EWL) was
59±15 at 2 years and 56±16 at 3 years, and then progressive regain occurred; at 5 years %EWL was
46±22 and at 10 years 40±27. Spearman rank test
failed to demonstrate any correlation between
weight loss at 5 years and patient data, except with
lifestyle score (Spearman r=0.8548, p<.0001). Current
mean age is 31±7 years.
Conclusion: BPD has to be considered for its
value in prolonging and qualitatively improving the
PWS patient’s life.
Key words: Prader-Willi syndrome, morbid obesity,
bariatric surgery, biliopancreatic diversion
Reprint requests to: Giuseppe M. Marinari, MD, Semeiotica
Chirurgica R, University of Genoa School of Medicine, Largo
Rosanna Benzi 8, 16132 Genova, Italy. Fax: +39.010.502.754;
e-mail: marinari@unige.it
© FD-Communications Inc.
Introduction
Prader-Willi Syndrome (PWS), described in 1956,1
is a congenital disorder associated with a partial
deletion of the long arm of chromosome l5,2 characterized by neonatal hypotonia, short stature,
hypogonadism, mental retardation of mild to
severe degree and compulsive hyperphagia with
the development of early obesity. The syndrome is
believed to occur with a frequency of one in 5,000
to 10,000 births. The obesity-related co-morbidities, such as respiratory failure with cor pulmonale,
diabetes mellitus, arteriosclerosis and its consequences, lead to a very short life-expectancy of
PWS subjects: they usually die between 20 and 30
years of age. Food obsession and mental retardation cause the failure of any weight reduction program entailing the subjects’ co-operation; conventional dietary treatment, which is essentially based
on voluntary restriction of food intake, has proven
totally inadequate in obtaining substantial and prolonged weight loss in PWS patients. 3 Therefore, a
surgical method seems to be the most reasonable
therapeutic approach to obesity in PWS patients.
Since gastric restrictive surgery requires strong
patient compliance to achieve weight loss, this surgical approach generally fails in PWS. Miyata 4
reported a transient improvement of glucose
metabolism with no weight loss in a single case
after vertical banded gastroplasty (VBG). Of the
two VBG patients described by Mason,5 one did
not lose at all, while the other lost 30% of the initial excess weight at 5 years and afterwards
regained. Slightly better results are reported with
Roux-en-Y gastric bypass (RYGBP). Anderson6
submitted 10 PWS patients to RYGBP, six of
Obesity Surgery, 11, 2001
491
Marinari et al
whom had their pouch revised subsequently; 5
years after RYGBP, although they were back to
their mean original weight, an increase in height
resulted in a percent excess weight smaller than
preoperatively. The three patients reported by
Dietz7 reverted to their initial weight within 3 years
after RYGBP.
Surgical methods based on reduction of intestinal
absorption, where weight loss results are relatively
independent of the patient’s co-operation, should
therefore be the most appropriate approach to PWS
obesity. The two reports on the use of jejunoileal
bypass (JIB) in PWS8,9 do not allow assessment of
its effectiveness, due to only 1-year follow-up.
Because of the side-effects, JIB has been abandoned by the vast majority of surgeons.
Biliopancreatic diversion (BPD) yields excellent
weight loss and long-term weight maintenance10 in
the absence of the untoward effects of JIB.
Methods
BPD for surgical treatment of obesity consists of a
distal gastrectomy with a long Roux-en-Y reconstruction which, by delaying the meeting between
food and biliopancreatic juices, causes a reduction
of digestion and consequently of intestinal absorption of energy-rich aliments (Figure 1). The procedure, in a series of 2,316 patients operated on during a 23-year period, caused a mean permanent
reduction of about 75% of the initial excess
jejunum
ileum
Figure 1. Biliopancreatic diversion.
492 Obesity Surgery, 11, 2001
weight.11 The indefinite weight maintenance
appears to be due to the existence of a threshold
absorption capacity for fat and starch, and thus
energy. Beneficial effects, other than those consequent to weight loss and/or reduced nutrient
absorption, include permanent normalization of
serum glucose and cholesterol without any medication and on totally free diet in 100% of cases, both
phenomena being due to a specific action of the
operation.12,13
Fifteen patients with PWS (9 male, 6 female)
underwent BPD between June 1986 and February
1996. In all cases the PWS diagnosis, made at a
mean age of 6.5 years-old (6 mo-14 yr), was
achieved in a children’s hospital. Of the 15
patients, 9 underwent chromosome examination,
and in 6 of the 9 the deletion of the proximal part
of the long arm of chromosome 15 was found. In
the other 6, a clinical diagnosis was made.
According to the Holm criteria,14 at time of BPD
all subjects had a total score of ³8. Informed consent was obtained in each case from parents.
Patient data are reported in Table 1: preoperative
mean age was 21±5 years, mean weight 127±26
kg, mean excess weight 74±25 kg (corresponding
to mean 142±50 percent of ideal weight), mean
Body Mass Index (BMI, kg/m2) 53±10, and mean
waist/hip 0.97 (no difference between M and F). At
the time of operation, 4 subjects had hypercholesterolemia, 3 had type II diabetes mellitus and 4 had
hypertension. Indication for BPD was BMI >40
and, in three patients, >35 with concurrent metabolic complications. Mean follow-up is 8.5 (4-13)
years.
The patients were evaluated 1, 3, 6 and 12
months after BPD and on a yearly basis thereafter.
Complete blood chemistry was obtained in all
instances. Wechsler Adult Intelligence Scale Test
for I.Q. assessment was performed at the last
examination in each patient, with a mean score of
72±10. An arbitrary score based on lifestyle was
given to each subject: 1= at home with parents,
total inaction and near absence of interest in life;
2= at home participating in housework, irregular
attendance at specific reference center; 3= regular
job and /or regular attendance at specific center
(Table 2).
BPD for Prader-Willi Syndrome
Table. 1. Preoperative patient data
Patients
Sex
BMI
Age at
Operation
(years)
Age at
Diagnosis
IQ score
Waist/Hip
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
F
M
F
M
F
M
M
M
M
M
F
M
F
F
M
36
40
57
55
68
39
58
59
53
47
61
53
72
54
46
23
24
27
12
25
16
31
16
17
29
18
19
21
18
20
10 yrs
6 mos
12 yrs
6 yrs
14 yrs
2 yrs
5 yrs
3 yrs
8 yrs
10 yrs
15 yrs
6 mos
10 yrs
2 yrs
9 mos
59
80
–
60
84
73
80
80
–
56
70
–
77
–
74
–
–
–
–
–
115/120
142/144
127/140
115/131
129/119
135/140
137/145
152/170
130/140
144/128
Results
No perioperative complications were observed. All
subjects were chronically supplemented with oral
calcium (2 g/day) and parenteral vitamin D
(200,000 U/month), while vitamin A, B complex
and iron were supplemented when advisable. Mean
body weight (Table 3) showed percent excess
weight loss (%EWL) of 59±15 at 2 years and
56±16 at 3 years, and then a progressive regain: at
5 years %EWL was 46±22 and at 10 years 40±27,
with marked interindividual differences. According
Table. 2 Lifestyle score according to patient’s daily occupation and long-term weight loss
Patient
Lifestyle Score*
%EWL at 5 yrs
1
2
3
4
5
6
7
8
9
10
11
12
13
14
2
3
1
1
3
2
3
3
1
3
3
3
1
2
36
57
4
35
65
36
64
68
50
51
66
77
15
27
*1= at home with parents, total inaction and near absence of
interest in life;
2= at home participating in housework, irregular attendance at
specific reference center;
3= regular job and/or regular attendance at specific center.
Total
Cholesterol
184
175
180
245
227
187
191
189
185
240
194
180
192
179
269
mg/dl
mg/dl
mg/dl
mg/dl
mg/dl
mg/dl
mg/dl
mg/dl
mg/dl
mg/dl
mg/dl
mg/dl
mg/dl
mg/dl
mg/dl
Fasting
Glucose
192 mg/dl
88 mg/dl
97 mg/dl
91 mg/dl
101 mg/dl
168 mg/dl
84 mg/dl
81 mg/dl
87 mg/dl
173 mg/dl
92 mg/dl
89 mg/dl
94 mg/dl
88 mg/dl
98 mg/dl
to BAROS criteria 15 for weight loss, of the 14 subjects at 5 years, eight (57%) showed an excellent or
good result (%EWL ³50), three (21%) a fair result,
(%EWL 25 to 49), and three (21%) failed (%EWL
£24). Of the seven subjects with a 10-year follow
up, three showed good results, one a fair result and
three were failures (one died during his ninth postoperative year). Weight loss at 5 years was greater
in males than in females (49% vs 34%, not significant). Spearman rank test did not demonstrate any
correlation between weight loss at 5 years and preoperative age, initial per cent excess weight, preoperative BMI, preoperative waist/hip ratio, waist circumference alone, age at diagnosis, and IQ score,
while a correlation was found (Spearman r=
0.8548, p <.0001) between weight loss at 5 years
and lifestyle score.
Both type II diabetes mellitus and hypercholesterolemia had disappeared 1 month after surgery,
with no relapse during the whole follow-up period,
even in patients who failed weight control.
Hypertension was cured in all the four affected
patients within the first postoperative year. One
patient had recurrent protein malnutrition and 24
months after BPD underwent surgical revision with
elongation of the common limb, when the %EWL
was 84; after the elongation of the common limb,
protein nutrition was permanently normalized and
the patient stabilized with %EWL 55%, which is
still maintained at 9 years. Two subjects had an
incisional hernia and one developed severe bone
demineralization. One of the unsuccessful subjects
Obesity Surgery, 11, 2001
493
Marinari et al
Table 3. % EWL after BPD in 15 PWS patients
Patient
%EWL
%EWL
1 yrs
2 yrs
3 yrs
4 yrs
5 yrs
6 yrs
7 yrs
8 yrs
9 yrs 10 yrs 11 yrs 12 yrs 13 yrs
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
67
78
166
150
216
74
159
162
138
107
187
136
237
153
103
41
83
48
58
31
64
67
73
58
68
46
66
26
40
61
47
88
51
59
56
53
84
73
64
62
46
69
36
39
56
41
80
48
54
66
45
73
63
79
61
44
75
29
37
49
39
69
8
44
67
42
66
66
50
53
49
75
24
27
48
36
57
4
35
65
36
64
68
50
51
66
77
15
27
39
59
8
0
47
26
55
72
33
41
†
81
18
57
4
13
67
26
55
68
29
44
37
57
8
10
66
31
45
71
22
43
61
9
†
67
36
54
71
34
61
4
49
67
0
46
51
0
72
15
54
70
70
mean
142
55
59
56
48
46
42
38
39
49
40
46
42
46
51
48
%EW=percent of excess weight; %EWL= percent of excess weight lost
died 9 years postoperatively from respiratory failure, while another subject with a very good weight
loss died in the sixth year from causes unrelated to
either obesity or BPD. No other significant complications occurred during the postoperative
period.
Discussion
The BPD weight loss results are due to a reduction
of intestinal absorption of energy, which, due to the
anatomo-functional characteristics of the operation, cannot be substantially modified by intestinal
adaptation mechanisms. 16 The strict indefinite
weight maintenance is due to the fact that intestinal
absorption of energy is not a percentage of the
intake but has a mean maximum threshold corresponding to about 1600 cal/day, so that body
weight after BPD is essentially independent of
food intake.17 Brossy18 reported good weight loss
results at 3 years in one PWS patient submitted to
BPD. The three cases described by LaurentJaccard19 did not show a consistent weight loss, but
they succeeded in maintaining a body weight less
than preoperative during a 5-year period. In a
recent paper by Antal,20 one patient had lost 80%
of the original excess weight 2 years after BPD,
while another patient had lost only 34% 1 year
after operation.
494 Obesity Surgery, 11, 2001
When considered as a group, the 15 PWS
patients in our series showed a satisfactory mean
weight loss 1 year after BPD, a result which was
maintained until the third year. When the patients
are considered individually, in the first 3 years all
of them lost weight. Unfortunately, an average progressive weight regain occurred afterwards. BPD
appears to cause a satisfactory weight loss in PWS,
but it cannot guarantee long-term weight maintenance in all operated PWS patients, unlike in the
general morbidly obese population.
A better weight loss result in males, could be due
to the greater genetically-determined percent fatfree mass despite hypogonadism. Nevertheless,
Schoeller18 found no differences in body composition between male and female PWS subjects, and
actually in our group there was no difference in
waist/hip ratio between males and females, indicating a distribution of adipose tissue irrespective of
gender; however, we did not find any correlation
between weight loss and waist/hip ratio or waist
circumference alone.
The limitation of energy absorption after BPD
concerns only the aliments that cannot be absorbed
in the absence of digestion, namely fat and starch,
while mono- and disaccharides, short-chain
triglycerides and alcohol are entirely absorbed.
Consequently, a certain degree of compliance in
avoiding excessive intake of sugar, sweets, fruit,
soft drinks, milk and alcoholic beverages is
BPD for Prader-Willi Syndrome
required from BPD patients to maintain the weight
attained. Therefore, an uncontrolled intake of simple sugar could explain the unusually poor weight
maintenance.
In addition, the continuous feeding, very common in PWS, could result in a more frequent saturation of the intestinal absorbent carriers and/or in
greater intestinal adaptation mechanisms: an
enhancement of the aforementioned energy
absorption threshold should ensue in both cases.
The surprising lack of correlation between IQ
score and long-term weight loss results shows how
predominant and compulsive hyperphagia is in
PWS. This magnifies the parents’ role: if they succeed in engaging their children in activities likely
to divert their interest in food for some time, they
will give them the chance to avoid a relapse in obesity.
In conclusion, although BPD provides the best
weight loss results for the treatment of obesity in
PWS patients, the procedure is often unable to
guarantee the long-term weight loss maintenance.
Nevertheless, taking into account the rapidly fatal
evolution of the disease, prolonging these patients’
life, while improving its quality, appears to recommend the use of this surgical procedure.
References
1. Prader A, Labhart A, Willi H. Ein Syndrome von
Adipositas, Kleinwuchs, Kryptorchismus, and
Oligophrenie nach myotonieartigem Zustand im
Neugeborenenalter. Schweiz Med Wochenschr
1956; 86: 1260.
2. Ledbetter DH, Riccardi VM, Airhart SD et al.
Deletions of chromosome 15 as a cause of the
Prader-Willi syndrome. N Engl J Med 1981; 304:
315-29.
3. Laurance BM, Brito A, Wilkinson J. Prader-Willi
syndrome after age 15 years. Arch Dis Child 1981;
56: 181-6.
4. Miyata M, Dousei T, Harada T et al. Metabolic
changes following gastroplasty in Prader-Willi syndrome: a case report. Jap J Surg 1990; 20: 359-64.
5. Mason EE, Scott DH, Doherty C et al. Vertical
banded gastroplasty in the severely obese under age
21. Obes Surg 1995; 5: 23-33.
6. Anderson AE, Soper RT, Scott DH. Gastric bypass
for morbid obesity in children and adolescents. J Ped
Surg 1980; 15: 876-80.
7. Dietz WH. Genetic Syndromes. In: Björntorp P,
Brodoff BN, eds. Obesity. JB Lippincott 1992: 58993.
8. Randolph JG, Weintraub WH, Rigg A. Jejunoileal
bypass for morbid obesity in adolescents. J Ped Surg
1974; 9: 341-5.
9. Rigg CA. Jejunoileal bypass for morbidly obese adolescent. Acta Paediatr Scand 1975; 256 (Suppl): 62-3.
10.Marinari G, Simonelli A, Friedman D et al. Very
long-term assessment of subjects with “half-half”
biliopancreatic diversion. Obes Surg 1995; 5: 124
(abstract 17).
11.Scopinaro N, Adami GF, Marinari GM et al.
Biliopancreatic diversion. World J Surg 1998; 22:
936-46.
12.Scopinaro N, Adami GF, Marinari G et al. The effect
of biliopancreatic diversion on glucose metabolism.
Obes Surg 1997; 7: 296-7 (abstract).
13.Marinari G, Adami GF, Camerini G et al. The effect
of biliopancreatic diversion on serum cholesterol.
Obes Surg 1997; 7: 297 (abstract).
14.Holm VA, Cassidy SB, Butler MG et al. Prader-Willi
syndrome: consensus diagnostic criteria. Pediatrics
1993; 91: 398-402.
15.Oria HE, Moorehead MK. Bariatric analysis and
reporting outcome system (BAROS). Obes Surg
1998; 8: 487-99.
16.Stock-Damgé C, Aprahamian M, Raul F et al.
Intestinal adaptation following biliopancreatic
bypass. Clin Nutr 1986; 5 (Suppl): 225-31.
17.Scopinaro N, Marinari GM, Camerini G et al.
Energy and nitrogen absorption after biliopancreatic
diversion. Obes Surg 2000; 10: 436-41.
18.Brossy JJ. Biliopancreatic bypass in the Prader-Willi
syndrome. Br J Surg 1989; 76: 313.
19.Laurent-Jaccard A, Hofstetter JR, Saegesser F et al.
Long-term results of treatment of Prader-Willi syndrome by Scopinaro’s biliopancreatic diversion:
study of three cases and the effect of dextrofenfluramine on the postoperative evolution. Obes Surg
1991; 1: 83-7.
20.Antal SC, Levin H. Biliopancreatic diversion in
Prader-Willi syndrome associated with obesity. Obes
Surg 1996; 6: 58-62.
21.Schoeller DA, Levitsky LL, Bandini LG et al.
Energy expenditure and body composition in PraderWilli syndrome. Metabolism 1988; 37: 115-20.
(Received January 7, 2001; accepted June 3, 2001)
Obesity Surgery, 11, 2001
495