Journal of

tUnical Pharmacy and Therapeutics

Journill of Clinical Phannacy and Therapeutics, 2012, 37, 187-195

Effects of I-year orIistat treatment compared to placebo on insulin resistance

•
doi, 1O.1l1l/j.1365-271O.2011.01280.x

parameters in patients with type 2 diabetes

C. Derosa* MD PhD, A. F. G. Cicerot MD PhD, A. D'Angelo" BD PhD, E. FogaTio- MD and P. Maffioli" MD

~Department of IntCrlmi Medicine and Therapeutics, University of Pavia, Pavia, Italy, and tDepartment of lntemal Medicine, Agiug lind Kidney di.c,cuscs,
"G. Descovich" Atherosclerosis Study Cenfer, University of Bologna, Bologna, Italy

Received 29 April 201 L Accepted 17 May 2011

Keywords: body weight, homeostasis model assessment index, resistin, retinol-binding protein-4, visfatin

and endometrial, breast, prostate and colon cancers.l Intensive

SUMMARY
programmes aimed at reducing calorie intake2 and increasing
What is known a"d Objective: The behavioural approach is physical activity-1 have been shown to improve metabolic con-
usually slow and not always sufficient to achieve optimal tar- trol in obese diabetic patients. However, the behavioural
gets in weight and metabolic control in obese diabetic approach is usually slow and not always sufficient to reach the
patients, and a pharmacological treatment is often necessary. optimal targets of weight and metabolic control in such patients,
The aim of this study was to compare the effects of orlistat and a pharmacological treatment L<; often nccessary.4.:; After the
and placebo on body weight, glycaemic and lipid profile and recent withdrawal of sibutramine's licence because of an
insulin resistance in patients with type 2 diabetes. increased risk of heart attack and stroke observed in the Sibutra-
Methods: Two hundred and fifty-four obese, diabetic patients mine Cardiovascular Outcomes TriaI/i orlistat is the only mole-
were enrolled in this study and randomized to take orlistat cule licensed in Europe as an anti-obesity drug. Whereas
360 mg or placebo for 1 year. We evaluated at baseline and sibutramine is a norepinephrine and serotonin reuptake inhibi-
after 3, 6, 9 and 12 months body weight, waist circumference tor? orlistat is the first prescription treatment for obesity that
(WO, body mass index (BM!), glycated haemoglobin (HbA tc ), does not act as an appetite suppressant. It interferes with the
fasting plasma glucose (FPG), post-prandial plasma glucose action of gastrointestinal (Cn lipase and thereby reduces the
(PPG), fasting plasma insulin (FPI), homeostasis model assess- absorption of dietary fat. tI Orlistat binds to the active site of CI
ment insulin resistance index (HOMA-IR), lipid profile, lipase to block its activity. The enzyme is thus unable to break
retinol-binding protein-4 (RBP-4), resistin, visfatin and high- triglycerides (Tg), and as a result, an estimated 30% of ingested
sensitivity C-reactive protein (Hs-CRP). dietary fat passes through the GI tract unchanged. Orlistat has
Results and Discussion: We observed a significant reduction in mainly mild to moderate GI side effects that usually attenuate
body weight, WC, BMI, lipid profile, RBP-4 and visfatin in with continued treatment but may often be unacceptable to
the orlistat group but not in control group. Faster improve- patients.'" The drugs also show some pharmacokinetic interac-
ments in HbAw PPG, FPI, HOMA-IR, resistin and Hs-CRP tions that are rare but potentially relevant, with cyclosporin 10
were recorded with orlistat than with placebo. A similar and warfarin. ll Orlistat has been shown to be effective, when
decrease in FPG was seen in the two groups. Significant pre- combined with lifestyle changes, in reducing the incidence of
dictors of change in insulin resistance (HOMA-IR) were RBP- type 2 diabetes mellitus in a clinically representative obese
4 and resistin concentration in the orlistat group (r = -0·53, population. 12 -14 Orlistat also improves the management of type
P < 0'05, and r = -0'59, P < 0'01, respectively). 2 diabetes mellitus in patients treated with insulin L~ or metfor-
What is new and Conclusion: To the best of our knowledge, min.16
this is the first study investigating the effect of orlistat on The aim of this study was to evaluate the effects of l-year
insulin resistance and markers of inflammation. Orlistat treatment with orUstat compared to placebo not only on body
improved lipid profile and led to faster glycaemic control and weight and on glycaemic control but also on insulin resistance
insulin resistance parameters than the control, without any parameters, such as retinol-binding protein-4 (RBP-4), resistin,
serious adverse event. Orlistat also improved RBP-4 and visfa- visfatin, and on inflammatory parameters, such as high-sensitiv-
tin, effects not observed with placebo. ity C-reactive protein (Hs-CRP) in type 2 diabetic obese patients
treated with different oral hypoglycaemic agents or insulin.
WHAT IS KNOWN AND OBJECTIVE
MATERIAL AND METHODS
Overweight and obesity are conditions that substantially raise
the risk of morbidity from hypertension, dyslipidaemia, type 2
Study design
diabetes mellitus, coronary heart disease, stroke, gallbladder
disease, osteoarthritis, sleep apnoea and respiratory problems, This multicentre, randomized, double-blind, controlled study
was conducted at the Department of Internal Medicine and
Correspondence: Giuseppe Derosa, MD, PhD, Department of Inter- Therapeutics, University of Pavia (Pavia, Italy), and the 'G. Des-
nal Medicine and Therapeutics, University of Pavia, Pavia, P.le C covich' Atherosclerosis Study Center, Department of Internal
Golgi, 2 - 27100 Pavia, Italy. Tel.: +390382526217; fax: +390382 Medicine, Aging and Kidney diseases, University of Bologna
526259; e-mail: giuseppe.derosa@unipv.it (Bologna, Italy). The study protocol was approved at each site

© 2011 Blackwell Publishing Ltd 187

Orlistat on insulin resistance G. Derosa et a1.

by institutional review boards and was conducted in accordance Table 2. Antidiabetic agents before and during the study
...vith the Declaration of Helsinki and its amendments.
We enrolled 254 Caucasian patients with type 2 diabetes aged
Orlistat group Control group
~18 of either sex according to the European Society of Cardiol-
ogy and European Association for the Study of Diabetes Guide-
lines criteria,t7 obese [body mass index (BMO ~30 kg/m 2JHI and n 126 128
with uncontmlled type 2 diabetes mellitus [glycated haemoglo- OHA, /I ('k) 119 (94·4) 120 (93·7)
bin (HbA k ) >8'0%1 on therapy with different oral hypoglycae- Sulphonylureas, Il ('7~) 22 (18·5) 20 (16·7)
mic agents or insulin. Suitable palients, identified from review Glyburide 8 (36·4) 7 (35'0)
Glimepiride 12 (54'5) 10 (SO'O)
of case notes and/or computerized clinic registers, were con-
Glicluide 2 (9'1) 3 (15'0)
tacted by the investigators in person or by telephone. Biguanides, Il (%) 75 (63·0) 77 (64,2)
Patients were excluded if they had a history of ketoacidosis Metformin 75 (101}0) 77 (10(}0)
or had unstable or rapidly progressive diabetic retinopathy, Glinides, Il (%) 19 (16·0) 20 (l6·7)
nephropathy or neuropathy; impaired hepatic function (defined Repaglinide 17 (89'5) 16 (80'0)
as plasma aminotransferase and/or gamma-glutamyltransferase Nateglinide 2 (10·5) 4 (20'0)
level higher than the upper limit of normal (ULN) for age and :x-glucosidase inhibitors, Il(%) 28 (23·5) 26 (21'7)
sex], impaired renal function (defined as serum creatinine level Acarbosc 28 000·0) 26 (W(}O)
Thiazolidinediones, II ('k) 53 (44-5) 56 (46'7)
higher than the ULN for age and sex) or severe anaemia.
Pioglitazone 42 (79·2) 40 (71'4)
Patients with serious cardiovascular disease (CVD) (e.g. New
Rosiglitazone 11 (20·8) 16 (28'6)
York Heart Association class I-IV congestive heart failure or a Incretin-mimetics, Il ('l) ') (7-6) 8 (6'7)
history of myocardial infarction or stroke) or cerebrovascular Exenatide 9 (100·0) 8 (10(}0)
conditions within 6 months prior to shldy enrolment were DPP-4 inhibitors, /I ('X) 13 (10'9) 15 (12'5)
excluded. We also excluded patients with GI disorders or those Sitagliptin 9 (69·2) 10 (66·7)
who had major abdominal surgery within 6 months of study Vildagliptin 4 (30·8) 5 (33'3)
enrolment. Women who were pregnant or brea'>t-feeding or of Insulin, II (%) 14 (11·1) 13 (10'1)
child-bearing potential and not taking adequate contraceptive Analogue, II ('lr,) 9 (64·3) 10 (76'9)
Lispro 5 (55'6) 7 (70'0)
precautions were also excluded. All patients provided written
Glulisine 4 (44·4) 3 (30'0)
inform('(."l consent to participate.
Long-acting, Il (%) 5 (35·7) 3 (23-1)
At the beginning of the study and for the whole observational Glargine 3 (60·0) 1 (33,3)
period, patients were taking different antidiabetic drugs. The NPH 2 (40·0) 2 (66-7)
complete list of the antidiabetic drugs taken is reported in
Table 2, whereas the complete list of the other concurrent medi-
Data are expressed as Il or 'Yo.
cations is reported in Table 1. OHA, oral hypoglycilemic agents; DPP-4, dipeptidyl peptidase-4 inhibi-
tors; NPH, neutral protamine Hagedorn.
Treatments
Patients were assigned to receive in the randomized, double-
blind shldy, in addition to their current antidiabetic therapy, opaque, white capsules in coded bottles to ensure blinding.
orlistat 360 mg (orlistat group) or placebo (control group) for Randomization involved drawing of envelopes containing ran-
12 months. Orlistat and placebo were supplied as identical, domization codes prepared by a statistician. A copy of the code
was provided only to the responsible person performing the
statistical analysi<;. The code was only broken after database
Table 1. Concurrent medications during the study lock, but could have been broken for individual subjects in
cases of an emergency. Medication compliance was assessed by
counting the number of pills returned at the time of specified
Orlistat group Control group clinic visits. At baseline, we weighed participants and gave
them a bottle containing a supply of study medication for at
Concurrent medications, 1/ ('.:') 110 (87-3) 106 (82·8) least 100 days. Throughout the study, we instructed patients to
ACE-I 29 (26,4) 32 (30·2) take their first dose of new medication on the day after they
ARBs 32 (29'1) 31 (29·2) were given the study medication. A bottle containing placebo or
C dlcium-antagonists 16 (14'5) 17 (16·0) the study medication for the next treatment period Was given
II-blockers 7 (6·4) 8 (7'5) to participants each 3 months. At the same time, aU unused
Diuretics 14 (12,7) 17 (16'0) medications were retrieved for inventory. All medications were
Stdtins 50 (45'4) 48 (45·3)
14 (12,7)
provided free of charge.
Fibrallo's 12 (11·3)
Onwgd-:i 12 (10,9) 9 (8'5)
Acetylsalicylic acid 97 (88'2) 100 (94·3) Diet a"d exercise
Ticlopidine 7 (6·4) 5 (4·7)
Subjects began a controlled-energy diet (near 600 Kcal daily def-
icit) based on American Heart Association recommendations 19
Data arc expressed as means ± SO or Il and ',i.
P not significant.
that induded 50% of calories from carbohydrates, 30% from fat
Sm. st., Smoking status; Diab. dUL, diabetes duration; ACE-I, angiotensin- (6% saturated) and 20% from protein..<>, with a maximum choles-
converting enzyme inhibitors; ARBs, angiotensin receptor blockers. terol content of 300 mg/ day and 35 g/ day of fibre. Patients

o 2Ull Blilckwell Publishing Ltd Journal of Clinical Pharmacy and Therapeutics, 2012, 37,187-195
188
Orlistat on insulin resistance G. Derosa et al.

were not treated with vitamins or mineral preparations during measured by ErA kit obtained from Phoenix Pharmaceuticals,
the study. Standard dietary advice was given by a dietitian Inc.,. The intra- and interassay CsV were lOClr and <%,
and/or specialist doctor. The dietitian and/or specialist doctor respectively?l Hs-CRP was measured with use of latex-
periodically provided instruction on dietary intake recording enhanced immunonephelometric assays on a BN II analyser
procedures as part of a behaviour modification programme and (Dade Behring, Newark, DE, USA). The intra- and interassay
then later used the subject's food diaries for counselling. Indivi- CsV were 5·7% and 1'3%, respectively>12
duals were also encouraged to increase their physical activity
by walking briskly for 20~30 min, 3~5 times per week, or by
Statistical analysis
cycling. The recommended changes in physical activity through-
out the study were assessed at each visit using the subject's An intention-to--treat analysis was conducted in patients who
activity diary. had received 2:1 dose of study medication and had a subsequent
efficacy observation. Patients were included in the tolerability
analysis if they had received 2:1 dose of trial medication and
Assessments had undergone a subsequent tolerability observation. The null
Before starting the study, all patients underwent an initial hypothesis that the expected mean body weight, BMI, HbA 1C1
screening assessment that included a medical history, physical FPG, PPG, FPI, HOMA-IR, TC, LDL-c' HDL-c' Tg, RBP-4, ,esis-
examination, vital signs and a 12-lead electrocardiogram. We tin, visfatin and Hs-CRP change from baseline to the end of
evaluated at baseline and after 3, 6, 9 and 12 months the follow- 12 months of double-blind treatment did not differ significantly
ing parameters: body weight, BMI, HbA1Cl fasting plasma glu- among the treatments was tested using repeated-measures ana-
cose (FPC), post-prandial plasma glucose (PPC), fasting plasma lysis of variance and analysis of covariance. By considering as
insulin (FPI), homeostasis model assessment insulin resistance clinically significant a difff'rence of at least the lOVe compared to
index (HOMA-IR), total cholesterol (Tc), low-density lipopro- the baseline and an alpha error of U'OS, the actual sample size
tein cholesterol (LOL-C), high-density lipoprotein cholesterol was adequate to obtain a power higher than 0·80 for all mea-
(HOL-C), Tg. RBP-4, resistin, visfatin and Hs-CRP. sured variables (body weight, BMI, HbA 1CI FPG, PPC, FPI,
For tolerability assessment, all adverse events were recorded. HOMA-IR, TC, LOL-C, HOL-C, Tg, RBP-4, resistin, visfatin and
All plasmatic parameters were determined after a 12-h over- Hs-CRP). Continuous variables were compared by analysis of
night fast, with the exception of ppc, detennined 2 h after a variance (ANOVA). Intervention effects were adjusted for addi-
standardized meal. Venous blood samples were taken for all tional potential confounders (sex and smoking status) using
patients between OS'OO and 09'00. We used plasma obtained by analysis of covariance (ANCOVA). ANOVA was also used to assess
addition of NarEOTA, 1 mg/mL, and centrifuged at 3000g for the significance of differences within and between groups. The
15 min at 4 "c. Immediately after centrifugation, the plasma statistical significance of the indept>ndent effects of treatments
samples were frozen and stored at -SO °C for no more than on the other variables was determined using ANCOVA. A one-
3 months. All measurements were taken in a central laboratory. sample f-test was used to compare values obtained before and
BMI was calrulated as weight in kilograms divided by the after treatment; two-sample f-tests were used for between-group
square of height in metres. Glycated haemoglobin level was comparisons?1 Statistical analysis of data was performed using
measured by a high-performance liquid chromatography the Statistical Package for Social Sciences software version 14.0
method (DIAMAT; Bio-Rad, Richmond, CA, USA; normal (SPSS Inc., Chicago, IL, USA). Data are presented as mean ± -
values 4-2-6'2%), with intra- and interassay coefficients of varia- standard deviation (SO). For all statistical analyses, P < 0·05 was
tion (CsV) of <2%.20 Plasma glucose was assayed by glucose considered statistkally significant.
oxidase method (GOD/PAP; Roche Diagnostics, Mannheim,
Gennany) with intra- and interassay CsV of <2%.21 Plasma insu-
RESULTS
lin was assayed with Phadiaseph insulin radioimmuno assay
(RIA) (Pharmada, Uppsala, Sweden) using a second antibody to
Study sample
separate the free and antibody-bound 125 I-insulin (intra- and
interassay CsV 4·6% and 7'3%, respectively).22 A total of 254 patients were enrolled in the study and 234 com-
The HOMA-IR index was calculated as the product of basal pleted the study; 121 (51,7%) were allocated to the control group
glucose (mM) and insulin levels (/IU/mL) divided by 22'5.2..1,24 and 113 (4S'3%) to the orlistat group. There were 20 patients (10
TC and Tg levels were determined using fully enzymatic techni- men and 10 women) who did not complete the study, and the
ques2S •26 on a clinical chemistry analyser (HITACHI 737; Hitachi, reasons for premature withdrawal included side effects as flatu-
Tokyo. Japan). Intra- and interassay CsV were 1·0 and 2·1 for TC lence (two men in orlistat group after 3 months, one man in
measurement, and 0·9 and 2·4 for Tg measurement, respectively. orlistat group after 6 months and one man in control group
HDL-C level was measured after precipitation of plasma apo after 12 months), constipation (one woman in orlistat group
B-containing lipoproteins with phosphotungstic acid,27 and after 3 months and one woman in orlistat group after
intra- and interassay CsV were 1·0 and 1'9, respectively; LDL-C 9 months), abdominal pain (one woman in orlistat group after
level was calculated by the Friedewald formula?" RBP-4 was 3 months and one woman in orlistat group after 9 months),
measured using a RBP-4 (Human) enzyme immunoassay (EIA) fatty /oily evacuation (one man in control group after 9 months,
kit (Phoenix Phannaceuticals, Inc., Burlingame, CA, USA). The one woman after 6 months in orlistat group and two men in
intra- and interassay CsV were <5·0% and <14·0%, respectively.2'I orlistat group after 9 months), increased defecation (one man in
Resistin value was measured by a commercially available control group after 6 months and one woman in orlistat group
enzyme-linked immunoassay (ELISA) kit (BioVendor Laboratory after 12 months), faecal urgency (one woman in orlistat group
Medicine, Bmo, Czech Republic). Intra-assay CsV was 3·4% and after 12 months), malaise (one man in control group after
interassay CsV was 6'9%, respectively.3/) Visfatin levels were 9 months and one woman in orlistat group after 6 months), lost

© 2011 Blackwell Publishing Ltd Journal ofClillical Plwnnacy and Therapeutics, 2012, 37, lS7~195
189
Orlistat on insulin resistance c. Derosa et al.

to follow-up (one man in control group after 3 months and one change was observed in the control group. Furthennore, body
woman in control group after 9 months) and withdrawn of weight and BM! values obtained with orlistat were significantly
informed consent (one woman in control group after 3 months) lower than the values observed in the control group after
(Figure 1). 12 months (P < 0·05) (Tables 3 and 4).

Baseline characteristics Glycaemic parameters

There were no significant differences between the two groups at Glycated haemoglobin decreased after 12 months (P < 0·05)
baseline with respect to mean age (52 ± 5 years in the control compared to baseline in the control group, and after 6, 9 and
group and 53 ± 6 years in the orlistat group), duration of dia- 12 months with orUstat (P < 0·05, P < 0·01 and P < 0·001, respec-
betes (5 ± 3 years in the control group and 6 ± 4 years in the tively), but there were no significant differences between the
orlistat group) or the presence of concomitant diseases, such as two groups (Tables 3 and 4). In both groups, there was a
hypertension, hypercholesterolaemi3, hypertriglyceridaemia and decrease in FPC after 12 months (P < 0·05, for both) compared
combined dyslipidacmia. There was no significant difference to baseline, and no differences between the two treatments were
regarding concurrent medications (Table 1). recorded (Tables 3 and 4).
A significant improvement of PPG was observed after
12 month." (P < 0·05) in the control group and after 9 and
Body weigllt alld BMI
12 months (P < 0·05 and P < 0·01, respectively) with orlistat
We obs('rved a significant reduction in body weight and BMI (Tables 3 and 4), but there were no significant differences
after 9 and 12 months (P < 0·05 and P < 0·01, respectively) com- between the two therapies.
pared to baseline in the group treated with orlistat whereas no

254 patients randomised

128 patients randomised to placebo

1 1
13 patients lost to follow-up: 7 patients lost to follow-up:

• 3 males for flautulen~ • 1 male for flautulence

• 2 females for costipation • 1 female for insomnia
• 2 females for abdominal pain • 1 male for fatty/oily evacuation
• 2 males and 1 female for fattyioily • 1 male for increased defecation
evacuation • I male for malaise
• I female for increased defecation • I male and 1 female lost to
• 1 female for fecal urgency follow-up
• 1 female for malaise • 1 female for withdrawn of
infortllt!d consent

1 1
113 patients completed the study n 121 patients completed the study in
sibutramine group control group

234 patients completed the study and

were included in the analysis
Fig. 1. Study design

if:;· 2011 Blackwell Publishing Ltd 'oumal of Clinical Pllarmacy and TlJerapellfics, 2012, 37, 187-195
190
Orlistat on insulin resistance C. DcroSfl et al.

Table 3. Body weight, glycaemic profile and insulin resistance data during the study

Orlistat group Control group

Baseline 3 month 6 month 9 month 12 month Baseline 3 month 6 month I} month 12 month

126 122 119 115 113 128 126 125 122 121
"
Sex (M/F) 62/64 60/62 59/60 57/58 57/56 66/62 65/61 64/61 62/60 61/60
Sm_ st. (M/A 21/25 21/24 20/23 20/23 20/22 19/22 18/22 18/22 18/21 18/21
Weight (kg) 94·5 ± 9'6 92·7 ± 9·4 90-3 ± 8-4 88'1 ± 7-2~ 85'0 ± 5'9f 91·7 ± 8·7 90·8 ± 8·5 91·0±8·3 89·7 ± 8·0 89'1 ± 7·8
BMI (kg/m2) 33'1 ± 2'9 32·5 ± 2·3 31'6±1-8 30-8±IY 29-8 ± J-2f 32'5 ± 2-3 32-2 ± 2-1 31'9±2'0 31'8 ± 1·9 31'0 ± 1'8
WC(cm) 102·0 ± 6·0 lOt'0±5'S 99·0 ± 4·0 97·0 ± 3'S~ 95'0 ± 3'0i' 101'0 ± 5·5 100'0 ± 5'0 99-5 ± 4·5 99·5 ± 4·5 99·(l ± 4-0
HbAk (%) 8-4±1'4 8·1 ± 1·2 7-7 ± 0'9~ 7-3 ± 0'6t 7'0 ± 0'5§ 8·2 ± 1-3 g·O ± 1·2 7'9 ± 1-1 7-8±1'O 7'9 ± O'9~
FPC (mg/dU 136 ± 16 132 ± 14 129 ± 13 126 ± 12 121±11~ 133 ± IS 130 ± 14 127 ± 13 124 ± 12 120±10~

PPG (mg/dU 174 ± 24 170 ± 19 163 ± 17 156 ± 16~ 149 ± 13t 171 ± 20 167 ± 18 162 ± 17 158 ± 16 155 ± 15~
FPI (/lU/mD 22·8 ± 5·7 22·1 ± 5·2 21-3 ± 4-7 20-2 ± 4'5~ 19-3 ± 4·2 t 22·4 ± 5·3 22·0 ± 5'1 21-5±4·g 20-8 ± 4·6 20'1 ± 4Y
HOMA- IR 7·7 ± 4·2 7'0 ± 3'6 6·6 ± 3·4 6·1 ± 3-2~ 5-6 ± 2'8t 7-4 :t 3'9 7'1 ± 3'7 6'8 ± 3'5 6'4 ± 3'3 6'O±3'1~

Data are means ± SO.

~p < {}05 vs. baseline; tP < 0'02 vs. baseline; tP < 0·01 vs. baseline; §P < 0·001 vs. baseline .
.. p < (}OS vs. Control group.
Sm. s1.. Smoking status; BMI, body mass index; Wc, waist circumference; HbA le , glycated haemoglobin; FPC, fasting plasma glucose; PPC, post-prandial
pla~iIna glucose; FPI, fasting plasma insulin; HOMA-IR- homeostasis model assessment insulin resistance index.
Conversion factors: FPC, and PPG mg/dL (0,0555 ffiM); FPI JIU/mL (6·945 pM).

Table 4. Lipid profile data during the study

Orlistat group Control group

Baseline 3 month 6 month 9 month 12 month Baseline 3 month 6 month 9 month 12 month

N 126 122 119 115 113 128 126 125 122 121
Sex 1M/F) 62/64 60/62 59/60 57/58 57/56 66/62 65/61 64/61 62/60 61/60
Sm. st. (M/F) 21/25 21/24 20/23 20/23 20/22 19/22 18/22 18/22 18/21 18/21
TC (mg/dL) 220 ± 24 212 ± 17 207 ± 15 194 ± 11~ 186:t9+t 217 ± 21 210 ± 16 205 ± 13 206 ± 14 212 ± 17
LDL-C (mg/ dL) 153 ± 15 149 ± 11 144 ± 8 134 ± 7"' 126 ± 6t! 151 ± 13 145 ± 9 141 ± 7 144 ± 8 149 ± 11
HDL-C (mg/ dLJ 45 ± 7 44 ± 6 46 ± 8 45 ± 7 46 ± 8 46 ± 8 45 ± 7 46 ± 8 44 ± 6 45 ± 7
Tg (mg/dU 109 ± 48 95 ± 39 84 ± 30 76 ± 27 72 ± 25~ 99 ± 41 102 ± 45 92 ± 37 90 ± 35 88 ± 32

Data are means ± SO.

• p < 0'05 vs. baseline; tP < 0·02 vs. baseline.
!P < 0·05 vs. Control group.
Sm. st., smoking status; TC, total cholesterol; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; Tg, trigJycerides.
Conversion factors: re, LDL-C, and HDL-C mg/dL (0'0259 mM), Tg mg/dL (0,0113 mM).

Fasting plasma insulin improved after 12 months in the con- was not significantly different to that seen with the control
trol group (P < 0·05) and after 9 and 12 months with orlistat (Tables 3 and 4).
(P < 0·05 and P < 0'02, respectively) (Tables 3 and 4, Figure 2b).
Insulin resistance parameters
Lipid profile
A significant decrease in HOMA-IR was seen after 12 months in
No variations of lipid profile were recorded in the control the control group (P < 0·05) and after 9 and 12 months (P < 0·05
group, whereas an improvement of TC and LDL-C was and P < 0·02) with orlistat, but with no significant difference
observed with orlistat after 9 and 12 months (P < 0·05 and between the two groups (Tables 3 and 4, Figure 2a). We
P < 0'02, respectively). Furthermore, the values obtained with observed an improvement in resistin after 12 months in the
orlistat were significantly better than those observed in the con- control group (P < 0·05) and after 9 and 12 months with
trol group after 12 months (P < 0·05). Orlistat also improved Tg orHstat (P < 0·05 and P < 0·02, respectively) but no Significant
after 12 months (P < 0·05) compared to baseline, but the effect difference between the two groups (Tables 3 and 4, Figure 2d).

© 2011 Blackwell Publishing Ltd Jourllal of Clinical Phamwcy and Thera/Jeutics, 2012, 37,187-195
191
OrlL<;tat on insulin resistance C. Derosa et at.

(a) 10 _____ 0 (b) c 26

~ --+-0
.5 ....
e: -:)-",c
"E ...E

~
R

~
.t
~ ;~ 22
~
c c.~
6 ~-
= • t
_5
'; • t
4
0 3 6 9 12 '" 1"
0 3 6 9 12
Months Months
(C) (d)
~ _____ 0
-,,i _
c 46·0
:::;
8-0
~o

-• ~ ~
7-0
~sl
~~
c "-
:E"-' ,\4-0
"'0-0
i.!i 6-0

• ••t :'i·a ~
t
"i
" 2g-0
() 3 6 9 12
'" 4-0
0 3 6 9 12
l\'lonths :\1onths

(e) ]<)-0 (I) 3-0

-+-p
~o
:::;
~

~
lX-O 2·5
.§ ~c

S J7·{j
.§ ]6·[)
.
g
"'-I• 1·5
2·0

.5
:~ •
~
15-0
14-lJ
* = 1-0 t
Fig. 2. Inflammatory and insulin
0 3 6 9 12
" 0
Months
9
" Month.'!
resistance parameters variations during
the study.

A reduction of RBP-4 and visfatin was obtained after 12 months reduction in BMI, waist circumference, FPC, PPC, HOMA-IR,
(P < 0·05, for both) with orlistat, but not in the control group. levels of LDL-C, Tg, lipoprotein (a) and apolipoprotein B, uric
However, no Significant diffen'nce was observed betwecn thc acid level, pulse rate and systolic and diastolic blood pressure,
two treatments (Tables 3 and 4, Figures 2c and 2c). with a consequent reduction in CVD risk profile.:t5 Also, our
group has conducted several studies about the effects of orlistat,
both alone16-_19 or in association with L-carnitine.4t),41 We
InflammatonJ state
showed that orlistat reduced TC, LDL-C and Tg significantly in
There was an improvement of Hs-CRP after 12 months hypercholesterolaemic and obese patients-16.-17 and blood pres-
(P < (JoOS) in the control group, and after 9 and 12 months with sure in hypertensive obese patientsJ8 and was more effective
urlistat (P < 0·05, and P < 0·02, respectively), but the effect was than placebo in ameliorating inflammatory parameters, such as
not significantly different between the two therapies (Tables 3 adiponectin and tumour necrosis factor-a.3'I Furthermore, orli-
and 4, Figure 20. stat plus L-carnitine improved body weight and glycaemic and
lipid profile better than orlistat alone and produced faster and
better improvement in insulin resistance,40 and inflammatory
CORRELATIONS
parameters.4 !
Stepwise multiple regression analysis was undertaken to estab- In the current study, we showed that only the group treated
lish which anthropomehic and metabolic factors could best pre- with orlistat had a reduction in body weight and BM!. It can
dict improvement in insulin resistance (HOMA-IR). Significant appear surprising that orlistat did not have any significant effect
predictors of change in insulin resistance (HOMA-IR) were on weight loss before 9 months and that its effects were not sig-
RBP-4 and resi<;tin concentration in orlistat group (r = -0-53, nificantly different from placebo before 12 months. This is prob-
P < 0·05 and r = -0'59, P < 0-01, respectively). Other correlation ably due to the fact that both groups underwent an intensive
analysis did not indicate any other association. lifestyle intervention programme, which probably attenuated
the effects of orlistat. In fact, the Look Action for Health in Dia-
betes trial already proved that intensive lifestyle intervention
DISCUSSION itself can produce sustained weight loss and improvements in
Several studies have already reported the positive effects of orli- fitness, gIycaemic control and cardiovascular risk factors in indi·
stat on body weight. J4 •J5 A previous comparative study carried viduals with type 2 diabetes mellitus. 42 We also observed a bet-
out on female patients showed that orlistat induced a significant ter and faster improvement of TC and LOL-C with orlistat

1;., 2011 Blackwdl Publishing Ltd JOlin/at of Clil1ical Pharmacy alld Thcrapcldics, 2012, 37, 187-195
192
Orlistat on insulin resistance G. Derosa et a1.

compared to placebo, whereas no differences between the two nounced than in simply obese patients. This may explain the
groups were observed regarding HDL-C and Tg, confirming similar effects of orlistat and placebo on insulin resistance.
what has already been reported by US. 36- 19 The results are also In our study, orlistat gave a faster improvement than placebo
in line with what those reported by Hutton and Fergusson41 in Hs-CRP, which has been shown to independently predict
and Lindgarde. 44 This is probably in part because of the myocardial infarction, stroke and peripheral artery disease."i4
decreased absorption of dietary Tg as a result of inhibition of Some positive effects of orlistat seem to be independent of
intestinal lipases by orlistat and in part because of the orlistat's weight loss55 as even after adjusting for percentage of body fat
independent effect on cholesterol absorption.4~ Orlistat was also and waist circumference, the decrease in serum leptin and the
faster than placebo in improving glycaemic profile, an effect increase adiponectin were significantly different \vith orlistat
consistent with earlier reports.'tM 7 than with placebo. Orlistat enabled effective management of
With respect to insulin resistance, literahIre reports suggest obeSity-related comorbidities, especially insulin resistance and
that patients with type 2 diabetes have higher HOMA-IR than atherosclerosis risk.
subjects with normal glucose tolerance subjects. 48 Our study Gastrointestinal effects were more common in the orlistat
showed that orlistat improved HOMA~IR faster than the control, group than in the control group although the difference was not
in line with results reported by Gokcel ct alY' statistically significant. The reported adverse effects were flatu-
Retinol-binding protein-4 concentration has been reported to lence, constipation, abdominal pain, fatty loily evacuation,
be increased in subjects with obesity, insulin resistance or type increased defecation, faecal urgency and malaise. All the events
2 diabetes compared with lean subjects,49 but the mechanisms were reported as mild or moderate, in line with our previous
by which RBP-4 induces insulin resistance are not well under- studies?6-41,56 Orlistat intake was not associated with any cardi-
stood. Resistin is produced by mononuclear cells and activated ovascular effects and was generally well tolerated.
macrophages, and it has been demonstrated that overexpression Limitations of our study include not evaluating whether the
of resistin decreases the ability of insulin to suppress hepatic beneficial effects on glycaemic control, body weight, lipid profile
glucose output or increase glucose uptake by muscle:';1J Avail- and insulin resistance parameters were sustained after the cessa-
able data also suggest that resistin increases inflammation and tion of therapy. However, Kelley ef al. 15 have previously
atherosclerosis. 51 Visfatin was identified as a secretory protein observed that orlistat plus diet therapy produced greater weight
highly enriched in human visceral adipocytes and also loss and improvements in glycaemic control and serum lipid
expressed in the liver, muscle, bone marrow and lymphocytes, concentrations than placebo plus diet therapy and that these
where it was first identified as pre-B-cell colony-stimulaling fac- changes were sustained during 1 year of treatment. Another
tor.52,.~3 The expression and secretion of visfatin is increased limitation is that we only studied a limited number of insulin
during the development of obesity, but in contrast with inflam- resistance biomarkers.
matory cytokines, this increase does not decrease insulin sensi-
tivity. Instead, visfatin exerts insulin-mimetic effects in culhIred
WHAT IS NEW AND CONCLUSIONS
adipocytes, hepatocytes and myohIbes and lowers plasma glu-
cose in mice:52 Visfatin binds to the insulin receptor with similar To the best of our knowledge, this is the first study investigat-
affinity but at a site distinct from insulin.52 In contrast to insu- ing the effect of orlistat on insulin resistance and markers of
lin, visfatin levels do not change with feeding and fasting. 52 It inflammation. Orlistat improved lipid profile and led to faster
remains to be determined whether visfatin acts in concert with glycaemic control and insulin resistance parameters than the
insulin to regulate metabolism and whether such an interaction control, without any serious adverse event. OrJistat also
occurs via endocrine or paracrine mechanisms. In our shIdy, we improved RBP-4 and visfatin, effects not observed with placebo.
observed that orlistat, added to prior antidiabetic therapy, gave
a faster decrease in resistin in the orlistat group and an
ACKNOWLEDGEMENTS
improvement in RBP-4 and visfatin, not observed in the control
group. The lack of difference between the orlistat and placebo The authors certify that they have no affiliation with, or finan-
groups is probably a reflection of the short observational period. cial involvement in, any organization or entity with a direct
In patients with diabetes, insulin resistance is also more pro- financial interest in the subject matter or materials discussed in
the manuscript.

REFERENCES cardiovascular risk reduction. Americllll outcomes in overweight and obese sub-
Joumal of Cardiology, 2007;99:688-79B. jects. New Ellslalld Jmlnlfll of Medicille,
1. Expert panel on the identification, ('valua- 4. Bolen S, Feldman L, Vassy J el Ill. Sys- 2010;363:972-974.
tion, and treatment of overweight in tematic review: comparative effcrtivem'Ss 7. Arterburn DE, Crane PK, Veenstra DL
adults Clinical guidelines on the identifi- and safety of oral medications for type 2 The efficacy and saff'ty of sihutraminf' for
cation, evaluation, and treatment of over- diabetes mellitus. Annllis of In/emlll Medi- weight loss: a systematic review. Arcl1ir~'5
weight and obesity in adults: executive cil1(',2007;147:386-399. of h!ternal MedicilU', 2004;164:994-1003.
summary. Americall JOUr/wi of Clinical 5. Norris SL, Zhang X, Avenell A cl Ill. Effi- 8. Leung WY, Thomas GN, Chan jC, Tom-
Nutritiol1,1998;68:899-917. cacy of phannacotherapy for weight loss linson B Weight management and current
2. American Diabetes Association Physical in adults with type 2 diabetes mellitus: a options in pharmacotherapy: orlistat and
activity I exercise and diabetes diabetes meta-analysis. ArciliUt'S of [,llallal Mt'diciIlC, sibutramint'. Clinical rl1l'rllpCUlit"5,
care. Diabetes Cllre, 2004;27:58-62. 2004;164:1395-1404. 2003;25:58-80.
3. Lee M, Aronne LJ Weight management 6. James WP, Caterson TO, Coutinho W l'I al. 9. Wong NN, Cheng-Wi A Orlistat. Ileart
for type 2 diabetes mellitus: global Effect of sibutramine on cardiovascular Di5<'1151.',2000;2:174-181.

© 2011 Blackwell Publishing Ltd IOllrnal of Clinical Pharmacy and Therapeutics, 2012, 37, 187-195
193
Orlistat on insulin resistance G. Derosa et al.

10 Zhi j, Moure R, Kanitril L, Mulligan TE 21. European Diabetes Policy Group A desk- 34. Davidson MH, Hauptman J, Di Girolamo
Phannacokinctic evaluation of thl' possi. top guide to type 2 diabetes mellitus. Dia- M et al. Weight control and risk factor
ble interaction between selected con· bt'lic Ml'dicil1l.', 1999;16:716-730. reduction in obese subjects treated for
comit<lnt medications and orlistat at 22. Heding LG Determination of total serum 2 years with orlistat. A randomized con-
~tI!"dy state in healthy subjects. Jour/wi of insulin (lRI) in insulin-treatt>d diabetic trol1ed trial. Journal of flit' American Medical
c/i1liclIl PhllrlllllCO/OXY, 2002;42:1011-1019. patients. Dillbdologia, 1972;8:260-266. Association, 1999;281:235-242.
11. MilcWillter RS, Ffil,ser HW, Armstrong 23. Matthews DR, Hosker JP, Rudenski AS, 35. Gokcel A, Gumurdulu Y, Karakose H
KM Orli"tat enhances warfarin effect t'.:aylor BA, Treacher OF, Turner RC ef al. Evaluation of the safety and efficacy
AI/Ilals of Pharmacotherapy, 2003;37:510-512. Homeostasis model assessment: insulin of sibutramine, orlistat and metformin in
12. Torgerson J5, Hauptman L Boldrin Ml\, resistance and beta-cell function from fast- the treatment of obesity. Diabetes Om'Sily
Sjostrom L XENical in the prevention of ing plasma glucoSt., and insulin concentra- IIl1d Melabolism, 2002;4:49-55.
diabetes in obese subjeds (XENDOS) tions in man. Diabetologia, 1985;28:412--419. 36. Derosa G, Mugellini A, Ciccarelli L et I!/.
study: d randomized study of orlistat as 24. Wallace TM, Levy Je M<lUhews DR Use Effects of orlistat, simvastatin, and orlistat
an adjunct to lifestyle changes for the pre- and abuse of HOMA modeling. Diahett's + simvastatin in obese patients with
vention of type 2 diabetes in obese Cllre, 2004;27:1487-1495. hypercholesterolemia: a randomized,
patients. Dialldl's Clm', 2004;27:155-161 25. Klose S, Bomer K Enzymatische Bes- open-label trial. Current therapeutic
13 Rkhelsen B, Tonstad S, Rossner 5 l'f al. timmung des Gesamtcholesterins mit dem resl'llrcil, clinical alld experimental,
Effect of orlistat on weight regain and car- Greiner Selective Analyzer (GSA II). jour- 2002;63:621-633.
diova~ular rbk fdctors following a very- IW/ of Clinical Cht'mistry & Clilliml Bior/It'm- 37. Derosa G, Mugellini A, Ciccar€'lli L,
low-t'ncrgy diet in abdominally obt.'se iSlry, 1978;15:121-130. Fogari R Randomized double-blind, pla~
patients: a 3-year r,mdomized, placebo- 26. Wahlefeld AW (1974) Methods of t'1Izynllltic cebo-controlled comparison of the action
controlled study. Diabell's Carr, 2007;30:27- allalysis: Iriglycerides delamillalion after of orlistat, fluvastatin, or both on anthro-
32. L'nzymalic hydrolysis, 2nd English cdn. pometric measurements, blood pressure,
J.l. Heymsfield SB, Segal KR, Hauptman J l\ew York: Academic Press, Inc, 18-31. and lipid profile in obese patients with
1'1 al. Effects of weight lo~s with orlistat on 27. Havel RJ, Edr HA, Bragdon JH The dis- hypercholesterolemia prescribed a stan-
glucose tok·rilncc and progression to type tribution and chemical composition of dardized diet. Clinical Therape1./tics,
2 di<lbctt'S In obese adults. /lrchiPf.'S of ultracentrifugaUy separated lipoproteins 2003;25:1107-1122.
hllcnwl Medicilll', 20(XJ;160:1321-1326. in human serum. IOllnlal of Clinical lIm:sli- 38. Derosa G, Cicero AFG, Murdolo G 1'1111.
15 Kellt'y DE, Bray GA, Pi-Suny~'r FX 1'1 al galiml, 1955;34:1345-1353. Efficacy and safety comparative evalua~
Clinical efficacy of orlistat therapy in 28. Fricdcwald WT, Levy RI, Fredrickson OS tion of orlistat and sibutramine treatment
overwt'ight and obese patit'nts with insu- Estimation of the concentration of low in hypertensive obese patients. DiawfI's
lin-treated type 2 diabetes: a I-year rando- density lipoprotein in plasma, •.....ithout use Obesity and Metabolism, 2005;7:47-55.
mined controlled trial. Via/>etes Care, of the preparative ultracentrifuge. Clinical 39. Derosa G, Maffioli P, Sal ...adeo SAT ct ai.
2(X)2;25:1 033-1 041. Clil'mislry, 1972;18:499-502 Comparison of orlistat treatment and pla-
16. Miles jM, Leiter L, Hollander I' t'f al. 29 Takebayashi K, Suetsugu M, Wakabayashi cebo in obese type 2 diabetic patients.
Effect of orlistat in overweight and obese S, Aso Y, lnukai T Retinol binding pro- Expt'rt Opinioll 0/1 Pharmacothcrapy,
patients with type 2 diabetes treated with tein-4 and clinical features of type 2 dia- 2010;11: 1971-1982.
metformin. Diabl.'fes Carl', 2002;25:1123- betes patients. IOllrlwl of Clil/ieal 40. Derosa G, Maffioli P, Ferrari I I't al. Orli-
1128, End()crillOlogy & Metabo/ism, 2007;92:2712- stat and L-carnitine compared to orlistat
17 Ryden t, Standi E, Bartnik M l'f al. Guide- 2719. alone on insulin resistance in obese dia-
lines on diabdes, pre-diabeks, and cardi- 30. Yannakoulia M, Yiannakouris 1\:, Blfrher S, betic patients. Endocrine IOl/mal,
oVilo,cular di~t'ases: executive summary. Matalas AL, Klimis-Zacas D, Mantzoros 2010;57:777-786.
The Ta~k Forct' on Diabetes and Cardio- CS Body fat m<lSS and mdcronutrient 41. Derosa G, Maffioli P, Ferrari I et al. (2010)
vascular Diseases of the Euro)X'an Society intake in relation to circulating soluble Comparison between orlistat plus L-carni-
of Cardiolugy (ESC) and of the European leptin receptor, free leptin index, adipo- tine and orlistat alone on inflammation
Associfltion for the Study of Diabetes nectin, and resistin concentrations in heal- parameters in obese diabetic patients. Flw-
(EASD). ElImpcal/ Hearl JUI/rnal, thy humans lourlwl of Clil/leal di/menlal Clinical Plwrmacology, doi:
2()()7;28:88 .. 136. Endocrinology & Metabolism, 2003;88:1730- 10.1111/ j.1472-8206.2010.00888.x
lR. World H",alth Organization (1997) Obesity 1736 42. Look AHEAD Research Group, Wing RR
l"el'l'nlillS lind !1111111lging till' gll/Imlepidemic 31. K6mer A, Garten A, BIGher M, Tauscher Long-term effects of a lifestyle interven-
RepIH'1 of WHO nmslIltalllJII 011 obesity. Gen- R, Kratzsch J, Kiess W Molecular char- tion on weight and cardiovascular risk
eva; WHO ilcteristics of serum visfatin and differ- factors in individuals with type 2 diabetes
I Y Lichtl'nskin AH, Appel Lj, Brands M ential detection by immunoassays. JOl/rn1l1 mellitus; four-year results of the Look
ct Ill. Summary of American Heart Asso- of Clinical Endocrinology [.,. Metabo/ism, AHEAD trial. Archivt's of 111/(711111 Medicine,
ciation Dit'! dnd Lifestyle Recommenda- 2007;92:4783--4791. 2010;170:1566-1575.
tions Revision 2006. Aril'rivscll'msis, 32. Rifai N, Tracy RP, Ridk('f PM Clinical effi- 43. Hutton B, Fergusson D Changes in body
Tilrolllbl)si..;, IIIld VaSClllar Biology, cacy of an automated high-sensitivity wl:'ight and serum lipid profile in obese
2[)06;26:2186-21 91. c-reactive protein assay. C/il/ieal Chemistry, patients treated with orlistat in addition to
20 Bunn HF, Gabbay KH, Gallop PM Tht;> 1999;45:2136-2141. a hypocaloric diet: a systematic review of
glycosylation of haemoglobin. Relevance 33. Winer HJ (1971) Sialistimi principit's ill randomized elinical trials. AmeriCiIll lour-
to didbdes mellitus. SciC/1ce, 1'::178;200:21- ('xprrllllt'lltal design, 2nd edn. New York: lIal of Cll1l/cal Nutritioll, 2004;80:1461-1468.
27 Mt..'Graw-Hill.

£. 2011 Blackwell Publishing Ltd jOllmal of Clinical Pharmacy alld Therapeutics, 2012, 37, 187-195
1~4
Orlistat on insulin resistance G. Derosil et al.

44. Lindgarde F The effect of orlistat on body Study Group. Ob.:'sily Research, 2000;8:49- 52. Fukuhara A, Matsuda M, ]\;isnizawa M
weight and coronary heart disease risk 61. cI al. Visfatin: a protein Sl'crded by visc-
profile in obese patients; the Swedish 48. Li YB, Zhu DL, Tian HM et al. Character- eral fat that mimics the effects of insulin.
Multimorbidity Study. joumal of Intema/ istics of dysfunction of islet beta-cell in 5ciel1[1',2(XJ5;307:426--430.
Medicil1e, 2000;248:245-254. newly diagnosed type 2 diabetic patients. 53. Hug C, Lodbh HF Visfatin: a new aJipn-
45. Mittendorfer B, Ostlund RE Jr, Patterson Zhong/lila Yi XIIt' Yi (lllIall Xuc Za ZiJi, kine. SciclI(c, 2005;307:366-3h7.
BW, Klein S Orlistat inhibits dietary cho- 2006;86:2537-2541. 54. Zwacka TP, Hombach V. TOf71.'wski J
lesterol absorption. Obesity Research, 49. Cho YM, Youn HS, Lee H ('/ al. Plasma C-reactive protein-medidted lipoprotein
2001;9:599--604. retinol-binding protl'in-4 concentrations uptake by macrophages. GrclilalioJ/,
46. Hollander PA, Elbein SC, Hirsch JB cla/. are elevated in human subjects with 2001 ;103:1194--1197.
Role of orlistat in the treatment of obese impaired glucose tolerance and type 2 55. Hsieh q, Wdng PW, Liu RT t'I al. Orlistat
patients with type 2 diabetes. A I-year diabetes. Oiabrles Care, 2006;29:2457-2461. for ohesity: benefits beyond weight loss.
randomized double-blind study. Diabefes 50. Steppan eM, Bailey 5T, Bhat 5 ef al. The DwbctL'S Research ilIld C/lIl/cal Pmcftu',
Care, 1998;21:1288-1294. hormone r('sistin links obesity to diabetes. 2005;67:78-83.
47. ROssner S, SjOstrom L, Noack R, Meinders Nlliurc (London), 2001;409:307-312. 56 Derosa G, Ciclc'1'o AF, Murdolo C, Ciccarelli
AE, Noscda G Weight loss, weight main- 51. Reilly MP, Lehrke M, Wolfe MI., Rohatgi L, fogari R Comparison of metabolic effl'Cts
tenance, and improved cardiovascular risk A, Lazar MA, Rader DJ Resistin is an of orlistat and sibutramine ITl'atment in
factors after 2 years treatment with orlistat inflammatory marker of atherosclerosis in Type 2 diabetic obese patients. Diah'les
for obesity. European Orlistat Obesity humans. CirculativlI, 2005;111:932--939. NzJfritiOI1 al1d Metabolism, 2004;17:222·-229.

© 2011 Blackwell Publishing Ltd !OIlYl/Il! of Cliltical Phllrmllcy lind Therapeutics, 2012, 37, 187-195
195