Vous êtes sur la page 1sur 7

Original article

Changes in insulin sensitivity and secretion after sleeve


gastrectomy
G. Casella1 , E. Soricelli1 , L. Castagneto-Gissey1 , A. Redler1 , N. Basso1 and G. Mingrone2,3,4
1 SurgicalSciences Department, Medical School ‘Sapienza’ University, and 2 Department of Internal Medicine, Catholic University of Rome, Rome, Italy,
3 Department of Diabetes and Nutritional Sciences, Faculty of Life Sciences and Medicine, King’s College, London, UK, and 4 Medizinische Klinik und
Poliklinik III, Universitätsklinikum Carl Gustav Carus an der Technischen Universität, Dresden, Germany
Correspondence to: Professor G. Mingrone, Catholic University, Largo A. Gemelli 8, Rome, Italy (e-mail: gmingrone@rm.unicatt.it)

Background: Sleeve gastrectomy is indicated for the treatment of obesity and related co-morbidity
including diabetes. The dynamic changes in insulin secretion and sensitivity after sleeve gastrectomy
are unknown.
Methods: Whole-body insulin sensitivity was measured by the euglycaemic hyperinsulinaemic clamp
technique, and insulin secretion by C-peptide deconvolution after an oral glucose tolerance test (OGTT),
before and 3, 6 and 12 months after sleeve gastrectomy in morbidly obese subjects. The time course of
glucagon-like peptide (GLP) 1, as a marker of insulin secretion following OGTT, was also assessed.
Results: Ten patients were included in the study. Median (range) baseline insulin sensitivity (M-value)
increased from 84⋅0 (20⋅2–131⋅4) mmol per kg per min at baseline to 122⋅8 (99⋅0–179⋅3) mmol per
kg per min at 12 months after surgery (P = 0⋅015). Fasting insulin sensitivity, measured by homeostatic
model assessment of insulin resistance, which represents a surrogate index of hepatic insulin resistance,
decreased from 3⋅3 (1⋅9–5⋅5) to 0⋅7 (0⋅5–1⋅1) mg/dl⋅𝛍units/ml (P < 0⋅001). Total insulin secretion,
measured as incremental area under the curve (AUC), after OGTT decreased from 360⋅4 (347⋅9–548⋅0)
to 190⋅1 (10⋅1–252⋅0) mmol/l⋅180 min at 12 months (P = 0⋅011). The AUC for GLP-1 increased from
258⋅5 (97⋅5–552⋅6) to 5531⋅8 (4143⋅0–7540⋅9) pmol/l⋅180 min at 12 months after sleeve gastrectomy
(P < 0⋅001). In multiple regression analysis, 51 per cent of the M-value variability was explained by GLP-1
secretion.
Conclusion: Sleeve gastrectomy improved insulin sensitivity and reduced insulin secretion within
6 months after surgery. Although there was a correlation between insulin sensitivity and bodyweight,
the major driver of the improvement in insulin sensitivity was GLP-1 secretion.

Paper accepted 23 September 2015


Published online 9 November 2015 in Wiley Online Library (www.bjs.co.uk). DOI: 10.1002/bjs.10039

Introduction of biliopancreatic diversion (20 per cent increase from


baseline after RYGB versus almost 100 per cent increase
Sleeve gastrectomy (SG) has become used widely for the after biliopancreatic diversion)6 – 8 , its major action seems
surgical treatment of obesity and related co-morbidity to be exerted on hepatic insulin sensitivity9 – 12 .
including diabetes1 – 3 . Schauer and colleagues1 showed Abbatini and co-workers3 reported that insulin sensitivity
that diabetes remission, defined by a glycated haemoglobin was restored in subjects with type 2 diabetes 1 year after
level below 6 per cent, was achieved in 24 per cent of SG or RYGB. In contrast, Kashyap et al.13 did not find any
patients who underwent SG compared with 5 per cent improvement 2 years after SG, but they used a surrogate
of those who received medical treatment. However, its marker of whole-body insulin sensitivity. Recently, Bradley
mechanism of action has not yet been elucidated. and colleagues12 reported that the degree of improvement
Roux-en-Y gastric bypass (RYGB) is known to improve in hepatic and peripheral insulin resistance was similar
glycaemic control mainly by increasing insulin secre- following SG or RYGB.
tion and stimulation of glucagon-like peptide (GLP) 1 Few studies have investigated insulin secretion after SG.
secretion4,5 . Although the effect of RYGB on whole-body Using the intravenous glucose tolerance test, it was demon-
insulin sensitivity is relatively small compared with that strated that the first phase of insulin secretion was restored

© 2015 BJS Society Ltd BJS 2016; 103: 242–248


Published by John Wiley & Sons Ltd
Changes in insulin sensitivity and secretion after sleeve gastrectomy 243

fully a few days after SG14 . Bradley and co-workers12 started at 6 cm from the pylorus and proceeded upwards
showed that the total insulin secretion rate after a mixed until the angle of His. The SG was created using a lin-
meal was decreased whereas β-cell function was greatly ear stapler, which was applied alongside a 48-Fr calibrating
improved after SG or RYGB accompanied by loss of 20 bougie. The resection line was made avoiding the critical
per cent bodyweight. However, the dynamic changes in area by resecting the fundus 1⋅5 cm from the angle of His.
insulin sensitivity and secretion over time following SG are The residual gastric remnant capacity was 60–80 ml.
unknown. Therefore, insulin sensitivity was investigated by
the euglycaemic hyperinsulinaemic clamp (EHC) method,
Homeostasis model assessment
and insulin secretion by C-peptide deconvolution after an
oral glucose tolerance test (OGTT) before, and 3, 6 and 12 Hepatic insulin resistance was measured by means of
months after SG. the homeostatic model assessment of insulin resistance
(HOMA-IR), which was computed as fasting plasma glu-
cose (mmol/l) times fasting serum insulin (munits/litre)
Methods
divided by 22⋅5, as described previously16 . Because fast-
Participants scheduled for SG were recruited from the ing plasma levels of glucose and insulin are used to calcu-
Department of Internal Medicine at the obesity outpatient late HOMA-IR, and the fasting glucose level is regulated
clinic, Catholic University of Rome, Italy. Inclusion cri- entirely by the liver, HOMA-IR is considered worldwide as
teria for the study were: age 30–65 years; and body mass a measure of hepatic insulin resistance.
index over 40 kg/m2 or at least 35 kg/m2 in the presence
of complications such as arthritis with walking difficulty,
Euglycaemic–hyperinsulinaemic clamp
sleep apnoea or obesity-induced physical problems inter-
fering with lifestyle. Exclusion criteria were: presence of Peripheral insulin sensitivity was evaluated by a 3-h EHC
diabetes mellitus, coronary heart disease and/or interven- procedure. This technique is currently the standard for
tions for such disease in the previous 6 months (myocardial insulin sensitivity measurement. After inserting an intra-
infarction, unstable angina, coronary artery bypass, surgery venous cannula into the dorsal vein of the hand for sam-
or coronary angioplasty), liver cirrhosis or end-stage renal pling arterialized venous blood, and another in the antecu-
failure; participation in any other concurrent clinical trial; bital fossa of the contralateral arm for infusions, the patient
other life-threatening, non-cardiac disease; pregnancy; or rested in the supine position for at least 1 h. Arterialized
inability to give informed consent. blood was obtained by inserting the hand into a heating pad
The study protocol was approved by the Institutional (60∘ C). Insulin was administered using decreasing boluses
Review Board and conducted according to the principles given in 10 min, followed by a continuous infusion rate
of the Helsinki Declaration, with all subjects providing of 40 munits per m2 per min (6 pmol per kg bodyweight
written consent. All tests were carried out from 30 to 7 days per min)17 . To maintain plasma glucose levels in a normal
before surgery, and at 3, 6 and 12 months after SG. range (fasting levels), 20 per cent glucose was infused and
blood glucose levels were measured every 5 min. Once a
steady state had been reached, usually 2–3 h after start-
Anthropometry
ing the clamp, the glucose infusion rate represented the
On the day of each test, bodyweight was measured and insulin-mediated glucose uptake by the body (M-value in
rounded to the nearest 0⋅1 kg by a beam scale. Height μmol per kg fat-free mass per min). Whole-body peripheral
was assessed and rounded to the nearest 0⋅5 cm using glucose utilization was calculated during the last 40 min of
a stadiometer (Holatin, Crosswell, UK). Body composi- the steady-state insulin infusion.
tion was assessed by dual-energy X-ray absorptiometry
(DEXA) using a Lunar Prodigy system (GE Healthcare,
Insulin secretion
Fort Edward, New York, USA). DEXA permits measure-
ment of fat-free mass and fat mass. The basal insulin secretion rate was computed after an
overnight fast, and after the OGTT, using the C-peptide
deconvolution model18 . For every mole of insulin secreted
Sleeve gastrectomy
by the pancreas, the pancreas also secretes 1 mole of con-
SG was performed according to the technique described by necting peptide (C-peptide) that is not extracted apprecia-
Gagner and colleagues15 with a few modifications2 . Divi- bly by the liver, and instead breaks down insulin. Using the
sion of the vascular supply of the gastric greater curvature deconvolution function it is possible to compute insulin

© 2015 BJS Society Ltd www.bjs.co.uk BJS 2016; 103: 242–248


Published by John Wiley & Sons Ltd
244 G. Casella, E. Soricelli, L. Castagneto-Gissey, A. Redler, N. Basso and G. Mingrone

Table 1 Anthropometric data, insulin sensitivity and secretion, and glucagon peptide 1 secretion before, and 3, 6 and 12 months after
sleeve gastrectomy in ten patients
After sleeve gastrectomy

Baseline 3 months 6 months 12 months

Weight (kg) 116⋅3 (99⋅0–136⋅7) 98⋅6 (88⋅1–12⋅6) 89⋅2 (82⋅0–107⋅0)† 81⋅0 (64⋅0–89⋅8)‡
BMI (kg/m2 ) 41⋅6 (38⋅1–45⋅2) 36⋅4 (33⋅0–41⋅5) 32⋅8 (29⋅2–53⋅3)‡ 28⋅7 (25⋅3–30⋅5)‡
Fasting glucose (mmol/l) 5⋅1 (4⋅1–6⋅0) 4⋅8 (4⋅2–5⋅3) 4⋅5 (3⋅8–4⋅9) 4⋅4 (3⋅9–4⋅8)
Fasting plasma insulin (pmol/l) 103⋅1 (50⋅2–155⋅0) 59⋅0 (35⋅1–78⋅0) 50⋅0 (24⋅4–62⋅1)† 25⋅5 (5⋅8–64⋅2)‡
HOMA-IR (mg/dl⋅μunits/ml) 3⋅3 (1⋅9–5⋅5) 2⋅0 (0⋅9–2⋅5)† 1⋅6 (0⋅7– 2⋅1)‡ 0⋅7 (0⋅5–1⋅1)‡
M-value (mmol per kg fat-free mass per min) 84⋅0 (20⋅2–131⋅4) 110⋅7 (61⋅2–178⋅2) 120⋅0 (91⋅4–181⋅1)* 122⋅8 (99⋅0–179⋅3)*
Plasma insulin at steady state (pmol/l) 499⋅1 (434⋅0–562⋅3) 468⋅0 (404⋅5–538⋅1) 441⋅5 (381⋅0–481⋅8) 405⋅0 (356⋅1–497⋅2)
Fasting insulin secretion (pmol per m2 per min) 423⋅8 (52⋅0–748⋅7) 542⋅9 (73⋅8–594⋅3) 346⋅8 (38⋅5–602⋅4) 393⋅0 (80⋅0–516⋅5)
Total insulin secretion (mmol/l⋅180 min) 360⋅4 (347⋅9–548⋅0) 192⋅0 (56⋅6–348⋅6) 178⋅6 (10⋅5–251⋅2) 190⋅1 (10⋅1–252⋅0)*
Total GLP-1 secretion (pmol/l⋅180 min) 258⋅5 (97⋅5–552⋅6) 7002⋅9 (3953⋅0–8992⋅5)‡ 5589⋅2 (4552⋅4–7040⋅0)‡ 5531⋅8 (4143⋅0–7540⋅9)‡

Values are median (range). Fasting insulin sensitivity as homeostatic model assessment of insulin resistance (HOMA-IR) was computed as fasting plasma
glucose × fasting serum insulin divided by 22⋅5. BMI, body mass index; GLP, glucagon-like peptide 1. *P < 0⋅050, †P < 0⋅010, ‡P < 0⋅001 versus baseline
(2-way repeated-measures ANOVA).

secretion from the circulating levels of C-peptide. The correlation coefficient. Multiple regression analysis was
total insulin secretion is the area under the curve (AUC) carried out to identify the determinants of insulin sen-
of the insulin secretion rate after the oral glucose load. sitivity. Probability plots were constructed to look for
normality of the residuals and lack of heteroscedasticity
Assays (uneven distribution of random disturbance, suggesting
non-linearity). Residuals were also assessed by use of the
For GLP-1 analysis, venous blood was collected in Durbin–Watson statistic. Casewise diagnostics did not
ice-chilled EDTA dipotassium-treated tubes containing reveal any obvious outliers. SPSS® version 13.0 (IBM,
400 × 106 units aprotinin per litre of blood. For all blood Armonk, New York, USA) was used for statistical analysis.
samples, the plasma was separated immediately by centrifu-
gation at 4∘ C and stored at −80∘ C pending assay. Plasma
Results
glucose was measured by the glucose oxidase method
(Beckman, Fullerton, California, USA). Plasma insulin Ten patients participated in the study, four men and six
was assayed using a microparticle enzyme immunoassay women with a median age of 35 (range 19–49) years.
(Abbott, Pasadena, California, USA) with a sensitivity of
1 μunit/ml and an intra-assay coefficient of variation (CV)
of 6⋅6 per cent. Total GLP-1 was measured by radioim- Insulin secretion
munoassay (Linco Research, St Charles, Missouri, USA);
One year after SG, the patients had lost 30 per cent of
intra-assay and interassay CVs were 9–14 and 11–20
their initial median weight (Table 1). Fasting insulin lev-
per cent respectively. This assay has 100 per cent speci-
els dropped from 103⋅1 (50⋅2–155⋅0) pmol/l at baseline
ficity for GLP-1(7–36), GLP-1(9–36) and GLP-1(7–37),
to 25⋅5 (5⋅8–64⋅2) pmol/l 1 year after SG (P < 0⋅001).
and does not cross-react with glucagon (0⋅2 per cent),
The fasting insulin secretion rate decreased from 423⋅8
GLP-2 (less than 0⋅01 per cent) or exendin (less than 0⋅01
(52⋅0–748⋅7) to 393⋅0 (80⋅0–516⋅5) pmol per m2 per min.
per cent).
The maximum decrease was achieved by 6 months after
surgery and the rate then remained stable up to 12 months.
Statistical analysis
Before surgery, plasma glucose concentrations peaked at
All data are expressed as median (range) unless specified 50–60 min after the oral glucose load and then decreased
otherwise. Data were examined for normal distribu- slowly; after surgery the peak came earlier at 30–50 min
tion according to Shapiro–Wilks criteria. Two-way after oral glucose loading, and glucose levels then declined
repeated-measures ANOVA with Tukey’s post hoc test rapidly to hypoglycaemic values of around 3⋅3–3⋅7 mmol/l
was used to analyse the effects of SG on study outcome at 120 min (Fig. 1). Later on hypoglycaemic symptoms
measures. The AUC for insulin secretion and GLP-1 developed in three of ten patients. Although plasma insulin
concentrations was calculated by the trapezoidal rule. levels were higher before surgery, after SG the peak value
Correlation analyses were performed using the Spearman was reached more than 70 min earlier (Fig. 2).

© 2015 BJS Society Ltd www.bjs.co.uk BJS 2016; 103: 242–248


Published by John Wiley & Sons Ltd
Changes in insulin sensitivity and secretion after sleeve gastrectomy 245

10 4000
9
3500

Insulin secretion rate (pmol/min)


8
Plasma glucose (mmol/l)

3000
7

6 2500

5 2000
4
1500
3 Before operation
Before operation
3 months 1000
2 3 months
6 months
1 6 months
12 months 500
12 months
0
20 40 60 80 100 120 140 160 180 0
20 40 60 80 100 120 140 160 180
Time (min)
Time (min)

Fig. 1 Mean plasma glucose concentration in the oral glucose


Fig. 3Insulin secretion rate time course before, and 3, 6 and 12
tolerance test carried out before, and 3, 6 and 12 months after
months after sleeve gastrectomy. The curves are fittings of the
sleeve gastrectomy. The dotted line indicates the lowest glucose
insulin secretion rate measured by the minimal mode
level (3⋅3 mmol/l). The glycaemic peak occurred at 60 min after
glucose loading before surgery compared with 30–50 min after
sleeve gastrectomy 200 Before operation
180 3 months
6 months
1000 Before operation 160 12 months
Plasma GLP-1 (pmol/l)

3 months
900 140
6 months
800 12 months 120
Plasma insulin (pmol/l)

700 100
600 80
500 60
400 40
300
20
200
0
100 30 60 90 120 150 180
Time (min)
0
20 40 60 80 100 120 140 160 180
Time (min) Plasma glucagon-like peptide (GLP) 1 concentration time
Fig. 4
course before, and 3, 6 and 12 months after sleeve gastrectomy.
Plasma insulin concentration time course before, and 3, 6
Fig. 2 Values are mean(s.d.)
and 12 months after sleeve gastrectomy. The peak occurred
between 90 and 120 min before sleeve gastrectomy, and at
baseline to 0⋅7 (0⋅5–1⋅1) mg/dl⋅μunits/ml at 12 months
30–60 min after surgery. Values are mean(s.d.)
after surgery (P < 0⋅001). Insulin sensitivity (M-value) mea-
sured by the hyperinsulinaemic clamp method increased
Total insulin secretion after the OGTT had decreased from 84⋅0 (20⋅2–131⋅4) to 122⋅8 (99⋅0–179⋅3) mmol per
by 47 per cent by the end of the study (Fig. 3 and Table 1), kg fat-free mass per min (P = 0⋅015) at 12 months after SG,
and its reduction followed the same trend as fasting insulin but it had already reached a plateau at 6 months (Table 1).
secretion. A similar trend was observed for GLP-1 secretion (Fig. 4),
which increased rapidly by 3 months after SG, declined
slightly by 6 months and stabilized thereafter (Table 1).
Insulin sensitivity
The AUC for GLP-1 after the OGTT increased from
The fasting insulin sensitivity measured by HOMA-IR, 258⋅5 (97⋅5–526⋅6) pmol/l⋅180 min at baseline to 5531⋅8
which represents an index of hepatic insulin resis- (4143⋅0–7540⋅9) pmol/l⋅180 min at 1 year (P < 0⋅001)
tance, decreased from 3⋅3 (1⋅9–5⋅5) mg/dl⋅μunits/ml at (Table 1).

© 2015 BJS Society Ltd www.bjs.co.uk BJS 2016; 103: 242–248


Published by John Wiley & Sons Ltd
246 G. Casella, E. Soricelli, L. Castagneto-Gissey, A. Redler, N. Basso and G. Mingrone

The insulin sensitivity M-value at 1 year, measured using the euglycaemic clamp method. Although GLP-1 may play
the euglycaemic clamp, correlated well with the total a role in the observed improvement in insulin sensitivity
GLP-1 secretion above baseline (R = 0⋅71, P < 0⋅001) and after SG, in the present study it explained only half of the
with bodyweight (R = −0⋅70, P = 0⋅001). However, in a M-value variability. Further studies are needed to clarify
stepwise regression model with M-value as the dependent the mechanisms involved, including the contribution of
variable, and GLP-1 AUC and bodyweight as independent factors other than GLP-1.
variables, weight was no longer significant, whereas 51 per It has been shown that chronic subcutaneous adminis-
cent (P < 0⋅001) of the M-value variability was explained by tration of GLP-1 increases insulin sensitivity in insulin-
GLP-1 secretion (GLP-1 standardized coefficient β = 0⋅71, resistant C57Bl/6 J mice fed a high-fat diet22 . In addi-
P < 0⋅001; weight standardized β = −0⋅34, P = 0⋅303). tion, there is evidence that GLP-1 and its analogues
improve insulin resistance in obese animal models23 and in
patients with type 2 diabetes mellitus24 . Recently, Green
Discussion
and co-workers25 showed that GLP-1 receptor protein
In this study, insulin sensitivity improved markedly is expressed in human myocytes and that, under eugly-
after SG and this was associated with increased serum caemic conditions, GLP-1 promotes glucose uptake by
GLP-1 levels. In addition, hypoglycaemia with levels of increased membrane expression of glucose transporter 4
3⋅3–3⋅7 mmol/l was elicited by simple carbohydrate inges- and enhances glycogen synthesis. Therefore, it is likely
tion, and probably occurred as a result of rapid delivery that the remarkable increase in GLP-1 secretion observed
of glucose from the stomach into the small intestine, after SG does account for the net improvement in insulin
stimulating a quick insulin response. sensitivity observed in the present study.
In previous studies, insulin sensitivity after SG was SG is associated with rapid gastric emptying and accel-
found to be either increased12,14 or unmodified19 . The erated small bowel transit of semisolids in animals26 and
variability in these results could be due to differences in humans27 . The accelerated intestinal transit, which seems
demographics of the study population, study design or to be partially independent of the faster gastric emptying27 ,
methods used to assess insulin sensitivity. The current might be responsible for the rapid increase in insulin secre-
standard for assessment of insulin sensitivity is the EHC tion and consequently explain the hypoglycaemic episodes
method, which was used in the present study. Insulin sen- observed. Interestingly, it has been reported that, although
sitivity increased progressively for up to 6 months after SG some patients reported hypoglycaemic symptoms and
and remained stable thereafter. The metabolic results at others did not, their gastric caloric emptying rate was
3 months after surgery may be affected by the low-calorie similar28 .
diet of liquid and soft foods that is usually prescribed for Hypoglycaemia has been described in one-third to
the first 2 months following SG, after which regular food one-half of patients underging SG29 . In the present study,
is introduced gradually. only three of ten patients had symptoms of hypogly-
The net improvement in insulin sensitivity correlated caemia, but all subjects exhibited plasma glucose levels as
highly with both GLP-1 and bodyweight. However, in low as 2⋅94 mmol/l (53 mg/dl). The American Diabetes
a linear multiple regression model bodyweight lost its Association30 established a plasma glucose concentration
influence on insulin sensitivity, whereas GLP-1 remained of 70 mg/dl or less as the criterion for the definition of
the only predictor, explaining about 51 per cent of the hypoglycaemia; this glucose level represents the glycaemic
sensitivity to insulin. In an experimental study20 , adminis- threshold for glucagon activation31 and it reduces sympa-
tration of the GLP-1 receptor antagonist exendin (9–39) thoadrenal responses to subsequent hypoglycaemia32 . The
prevented improvements in glucose and insulin responses high incidence of hypoglycaemia in the present study is
after a meal in obese rats that underwent RYGB or SG. similar to that found after RYGB using a continuous glu-
In humans, Jiménez and colleagues21 found that blocking cose monitoring system. Kefurt and colleagues33 reported
the action of GLP-1 by exendin (9–39) infusion in type that 75 per cent of patients had hypoglycaemic episodes
2 diabetic subjects, who had remission of diabetes after with glucose levels lower than 3⋅05 mmol/l (55 mg/dl) after
SG, led to decreased insulin secretion but only a limited RYGB. A similar proportion (72 per cent) was reported
deterioration in glucose tolerance after surgery. This sug- by Roslin and colleagues34 . The present finding highlights
gests that GLP-1 is not the only factor responsible for the the need to advise consumption of frequent and small
improvement in glucose homeostasis after SG. However, meals after SG, and to avoid simple carbohydrates.
in that study insulin sensitivity was evaluated using the It is important to note that the insulinotropic action
AUC of the glycaemic response to a meal rather than with of GLP-1 is negligible in patients with a plasma glucose

© 2015 BJS Society Ltd www.bjs.co.uk BJS 2016; 103: 242–248


Published by John Wiley & Sons Ltd
Changes in insulin sensitivity and secretion after sleeve gastrectomy 247

concentration below 4⋅3 mmol/l35 . Therefore, it has been 3 Abbatini F, Capoccia D, Casella G, Soricelli E, Leonetti F,
hypothesized that the hypoglycaemic episodes observed Basso N. Long-term remission of type 2 diabetes in
after SG are due to the rapid delivery of glucose into morbidly obese patients after sleeve gastrectomy. Surg Obes
Relat Dis 2013; 9: 498–502.
the duodenum, which stimulates a brisk insulin response,
4 Laferrère B, Teixeira J, McGinty J, Tran H, Egger JR,
rather than being mediated by GLP-1 hypersecretion. In
Colarusso A et al. Effect of weight loss by gastric bypass
healthy subjects an intragastric pressure above 2 mmHg
surgery versus hypocaloric diet on glucose and incretin levels
is able to open the pylorus, permitting gastric content to in patients with type 2 diabetes. J Clin Endocrinol Metab
flow into the duodenum36 . Nutrient delivery and absorp- 2008; 93: 2479–2485.
tion in the small intestine is increased after SG. The 5 Van der Schueren BJ, Homel P, Alam M, Agenor K, Wang
pressure in the gastric sleeve during saline infusion has G, Reilly D et al. Magnitude and variability of the
been reported to be 43 mmHg, compared with 34 mmHg glucagon-like peptide-1 response in patients with type 2
before surgery37 . Thus, the higher intrasleeve pressure diabetes up to 2 years following gastric bypass surgery.
may be related to the increased gastric emptying reported Diabetes Care 2012; 35: 42–46.
after SG27 . 6 Salinari S, Bertuzzi A, Guidone C, Previti E, Rubino F,
Recently, Seeley and co-workers38 stressed the link Mingrone G. Insulin sensitivity and secretion changes after
between bile acids and improvement in glucose disposal gastric bypass in normotolerant and diabetic obese subjects.
Ann Surg 2013; 257: 462–468.
after bariatric surgery. The effects of both RYGB and
7 Guidone C, Manco M, Valera-Mora E, Iaconelli A, Gniuli
SG on glucose disposal are greatly reduced in knockout
D, Mari A et al. Mechanisms of recovery from type 2
nuclear bile acid receptor (FXR) rodents. These findings diabetes after malabsorptive bariatric surgery. Diabetes 2006;
may result from the changes in gut microbiota following 55: 2025–2031.
bariatric surgery. In fact, gut microbiota reduces the bile 8 Salinari S, Bertuzzi A, Asnaghi S, Guidone C, Manco M,
acid pool size and composition. It is important in the future Mingrone G. First-phase insulin secretion restoration and
to analyse bile acid and microbiota changes after SG, and differential response to glucose load depending on the route
their interactions with glucose metabolism, in humans. of administration in type 2 diabetic subjects after bariatric
This study has some limitations. The sample size is surgery. Diabetes Care 2009; 32: 375–380.
small and the oral glucose load is not a proper physiolog- 9 Jørgensen NB, Jacobsen SH, Dirksen C, Bojsen-Møller KN,
ical stimulus, although it gives a more prompt and pre- Naver L, Hvolris L et al. Acute and long-term effects of
dictable glucose absorption than using a mixed meal. Sec- Roux-en-Y gastric bypass on glucose metabolism in subjects
with type 2 diabetes and normal glucose tolerance. Am J
ond, hepatic insulin sensitivity was assessed by means of
Physiol Endocrinol Metab 2012; 303: E122–E131.
a surrogate index, the HOMA-IR, instead of being mea-
10 Schauer PR, Kashyap SR, Wolski K, Brethauer SA, Kirwan
sured directly with 6,6-deuterated glucose infusion. Finally, JP, Pothier CE et al. Bariatric surgery versus intensive
follow-up was only 1 year. It would have been of interest medical therapy in obese patients with diabetes. N Engl J
to study these patients 2–3 years after surgery, given that Med 2012; 366: 1567–1576.
bodyweight often increases 1–2 years after SG. This could 11 Rubino F. Bariatric surgery: effects on glucose homeostasis.
have provided more interesting information on the mecha- Curr Opin Clin Nutr Metab Care 2006; 9: 497–507.
nisms by which insulin sensitivity is increased by SG, which 12 Bradley D, Magkos F, Eagon JC, Varela JE, Gastaldelli A,
some authors consider the best current bariatric surgery Okunade AL et al. Matched weight loss induced by sleeve
option39 . gastrectomy or gastric bypass similarly improves metabolic
function in obese subjects. Obesity (Silver Spring) 2014; 22:
2026–2031.
Disclosure 13 Kashyap SR, Bhatt DL, Wolski K, Watanabe RM,
Abdul-Ghani M, Abood B et al. Metabolic effects of bariatric
The authors declare no conflict of interest.
surgery in patients with moderate obesity and type 2
diabetes: analysis of a randomized control trial comparing
References
surgery with intensive medical treatment. Diabetes Care
1 Schauer PR, Bhatt DL, Kirwan JP, Wolski K, Brethauer SA, 2013; 36: 2175–2182.
Navaneethan SD et al.; STAMPEDE Investigators. Bariatric 14 Basso N, Capoccia D, Rizzello M, Abbatini F, Mariani P,
surgery versus intensive medical therapy for diabetes – Maglio C et al. First-phase insulin secretion, insulin
3-year outcomes. N Engl J Med 2014; 370: 2002–2013. sensitivity, ghrelin, GLP-1, and PYY changes 72 h after
2 Abbatini F, Rizzello M, Casella G, Alessandri G, Capoccia sleeve gastrectomy in obese diabetic patients: the gastric
D, Leonetti F et al. Long-term effects of laparoscopic sleeve hypothesis. Surg Endosc 2011; 25: 3540–3550.
gastrectomy, gastric bypass, and adjustable gastric banding 15 Gagner M, Gumbs AA, Milone L, Yung E, Goldenberg L,
on type 2 diabetes. Surg Endosc 2010; 24: 1005–1010. Pomp A. Laparoscopic sleeve gastrectomy for the

© 2015 BJS Society Ltd www.bjs.co.uk BJS 2016; 103: 242–248


Published by John Wiley & Sons Ltd
248 G. Casella, E. Soricelli, L. Castagneto-Gissey, A. Redler, N. Basso and G. Mingrone

super-super-obese (body mass index > 60 kg/m2 ). Surg Today motility after sleeve gastrectomy: a prospective study. Ann
2008; 38: 399–403. Surg 2013; 258: 976–982.
16 Matthews DR, Hosker JP, Rudenski AS, Naylor BA, 28 Burgerhart JS, van Rutte PW, Edelbroek MA, Wyndaele
Treacher DF, Turner RC. Homeostasis model assessment: DN, Smulders JF, van de Meeberg PC et al. Association
insulin resistance and β-cell function from fasting plasma between postprandial symptoms and gastric emptying after
glucose and insulin concentrations in man. Diabetologia sleeve gastrectomy. Obes Surg 2014; 25: 209–214.
1985; 28: 412–419. 29 Papamargaritis D, Koukoulis G, Sioka E, Zachari E,
17 DeFronzo RA, Tobin JD, Andres R. Glucose clamp Bargiota A, Zacharoulis D et al. Dumping symptoms and
technique: a method for quantifying insulin secretion and incidence of hypoglycaemia after provocation test at 6 and
resistance. Am J Physiol 1979; 237: E214–E223. 12 months after laparoscopic sleeve gastrectomy. Obes Surg
18 Polonsky KS, Given BD, Hirsch L, Shapiro ET, Tillil H, 2012; 22: 1600–1606.
Beebe C et al. Quantitative study of insulin secretion and 30 Workgroup on Hypoglycemia, American Diabetes
clearance in normal and obese subjects. J Clin Invest 1988; Association. Defining and reporting hypoglycemia in
81: 435–441. diabetes: a report from the American Diabetes Association
19 Garrido-Sanchez L, Murri M, Rivas-Becerra J, Workgroup on Hypoglycemia. Diabetes Care 2005; 28:
Ocaña-Wilhelmi L, Cohen RV, Garcia-Fuentes E et al. 1245–1249.
Bypass of the duodenum improves insulin resistance much 31 Mitrakou A, Ryan C, Veneman T, Mokan M, Jenssen T,
more rapidly than sleeve gastrectomy. Surg Obes Relat Dis Kiss I et al. Hierarchy of glycemic thresholds for
2012; 8: 145–150. counterregulatory hormone secretion, symptoms, and
20 Chambers AP, Jessen L, Ryan KK, Sisley S, Wilson-Pérez cerebral dysfunction. Am J Physiol Endocrinol Metab 1991;
HE, Stefater MA et al. Weight-independent changes in 260: E67–E74.
blood glucose homeostasis after gastric bypass or vertical 32 Cryer PE. Diverse causes of hypoglycemia-associated
sleeve gastrectomy in rats. Gastroenterology 2011; 141: autonomic failure in diabetes. N Engl J Med 2004; 350:
950–958. 2272–2279.
21 Jiménez A, Mari A, Casamitjana R, Lacy A, Ferrannini E, 33 Kefurt R, Langer FB, Schindler K, Shakeri-Leidenmühler S,
Vidal J. GLP-1 and glucose tolerance after sleeve Ludvik B, Prager G. Hypoglycemia after Roux-en-Y gastric
gastrectomy in morbidly obese subjects with type 2 diabetes. bypass: detection rates of continuous glucose monitoring
Diabetes 2014; 63: 3372–3377. (CGM) versus mixed meal test. Surg Obes Relat Dis 2015; 11:
22 Parlevliet ET, de Leeuw van Weenen JE, Romijn JA, Pijl H. 564–569.
GLP-1 treatment reduces endogenous insulin resistance via 34 Roslin M, Damani T, Oren J, Andrews R, Yatco E, Shah P.
activation of central GLP-1 receptors in mice fed a high-fat Abnormal glucose tolerance testing following gastric bypass
diet. Am J Physiol Endocrinol Metab 2010; 299: E318–E324. demonstrates reactive hypoglycemia. Surg Endosc 2011; 25:
23 Young AA, Gedulin BR, Bhavsar S, Bodkin N, Jodka C, 1926–1932.
Hansen B et al. Glucose-lowering and insulin-sensitizing 35 Nauck MA, Heimesaat MM, Behle K, Holst JJ, Nauck MS,
actions of exendin-4: studies in obese diabetic (ob/ob, db/db) Ritzel R et al. Effects of glucagon-like peptide 1 on
mice, diabetic fatty Zucker rats, and diabetic rhesus monkeys counterregulatory hormone responses, cognitive functions,
(Macaca mulatta). Diabetes 1999; 48: 1026–1034. and insulin secretion during hyperinsulinemic, stepped
24 Zander M, Madsbad S, Madsen JL, Holst JJ. Effect of hypoglycemic clamp experiments in healthy volunteers.
6-week course of glucagon-like peptide 1 on glycaemic J Clin Endocrinol Metab 2002; 87: 1239–1246.
control, insulin sensitivity, and beta-cell function in type 2 36 Tougas G, Anvari M, Dent J, Somers S, Richards D,
diabetes: a parallel-group study. Lancet 2002; 359: 824–830. Stevenson GW. Relation of pyloric motility to pyloric
25 Green CJ, Henriksen TI, Pedersen BK, Solomon TP. opening and closure in healthy subjects. Gut 1992; 33:
Glucagon like peptide-1-induced glucose metabolism in 466–471.
differentiated human muscle satellite cells is attenuated by 37 Yehoshua RT, Eidelman LA, Stein M, Fichman S, Mazor A,
hyperglycemia. PLoS One 2012; 7: e44284. Chen J et al. Laparoscopic sleeve gastrectomy – volume and
26 Chambers AP, Smith EP, Begg DP, Grayson BE, Sisley S, pressure assessment. Obes Surg 2008; 18: 1083–1088.
Greer T et al. Regulation of gastric emptying rate and its 38 Seeley RJ, Chambers AP, Sandoval DA. The role of gut
role in nutrient-induced GLP-1 secretion in rats after adaptation in the potent effects of multiple bariatric
vertical sleeve gastrectomy. Am J Physiol Endocrinol Metab surgeries on obesity and diabetes. Cell Metab 2015; 21:
2014; 306: E424–E432. 369–378.
27 Melissas J, Leventi A, Klinaki I, Perisinakis K, Koukouraki 39 Cohen R. Sleeve gastrectomy: the ideal option for metabolic
S, de Bree E et al. Alterations of global gastrointestinal surgery? Nat Rev Endocrinol 2013; 9: 623.

© 2015 BJS Society Ltd www.bjs.co.uk BJS 2016; 103: 242–248


Published by John Wiley & Sons Ltd

Vous aimerez peut-être aussi