Key words
curcuminoids - randomized controlled trial - non-alcoholic fatty liver disease - liver
- steatohepatitis
Introduction
Non-alcoholic fatty liver disease (NAFLD) is a common hepatic disease characterized
by fatty infiltration in hepatocytes [1]. The prevalence NAFLD ranges from 15–30% in the overall Western populations [2]
[3]
[4]
[5], while its prevalence increases to 50–90% in obese individuals, more than 70% in
type 2 diabetic patients and nearly 100% in those subjects with both conditions [6]. Excessive fat accumulation and insulin resistance appear to play important roles
in the pathogenesis of NAFLD [7]. NAFLD encompasses a spectrum of liver abnormalities starting from simple hepatic
steatosis, steatohepatitis, liver fibrosis and cirrhosis which can finally progress
to hepatocellular carcinoma [8]. Although pathogenesis of NAFLD is complex and remains uncertain, it has been hypothesized
that lipid deposition in the liver is the primary event involved and subsequently
a variety of secondary processes, such as oxidative stress, lipid peroxidation, inflammatory
response, hepatic fibrosis and apoptosis, are triggered [9]. Owing to the increased global rates of obesity, a condition that is strongly associated
with NAFLD, development of novel effective therapies with minimal side-effects to
control or slow the progression of this disease is required.
Curcumin is a yellow phenolic compound that was initially isolated from Curcuma longa L. (turmeric) rhizomes in 1 815 [10]. Curcumin has been shown to possess numerous pharmacological effects including antioxidant,
anticarcinogenic, anti-inflammatory, antidepressant, lipid-modifying, anti-arthritic,
properties have been described for this polyphenol [11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]. Numerous pharmacological properties of curcumin are due to the interaction of this
natural product with various molecular targets such as enzymes, receptors, transcription
factors, growth factors, cytokines, adipokines, and other important biomolecules [24]
[25]
[26]
[27]
[28]
[29]. Furthermore, this polyphenol improves insulin sensitivity [15]
[30]
[31] and mitochondrial function [32], suppresses adipogenesis [33], reduces inflammation [34]
[35], fibrosis [36], and oxidative stress [37]
[38]
[39], that are causal risk factors of NAFLD. More importantly, experimental studies in
rodents with diet-induced NAFLD have suggested that curcumin treatment can reduce
hepatic lipid accumulation and progression of NAFLD to non-alcoholic steatohepatitis
(NASH) [40]
[41]. In clinical practice, although there is evidence on the efficacy of curcumin supplementation
in lowering both plasma triglyceride and cholesterol levels [42]
[44]
[44]
[45]
[46], there has been no study targeting subjects with NAFLD. Hence, the aim of the present
study was to evaluate the efficacy and safety of supplementation with curcumin in
subjects with NAFLD.
Materials and Methods
Subjects
Studied subjects were selected from adults referring to the Gastroenterology and Hepatology
Clinic of the Baqiyatallah Hospital (Tehran, Iran) for whom a diagnosis of NAFLD (grades
1–3) was made according to liver ultrasonography. Exclusion criteria were pregnancy
or breastfeeding, NAFLD secondary to alcohol consumption, smoking, consumption of
hypoglycemic, hypolipidemic and anti-inflammatory medications as well as any drug
known to affect hepatic function, and presence of hepatitis, coronary, renal, pulmonary
and thyroid diseases. All patients received dietary and lifestyle advises before the
start of trial.
Eligible subjects were randomly (via alternative assignment to capsules bottles coded as A or B) allocated to the curcumin
(1 000 mg/day in 2 divided doses) (n=50) or control (n=52) group. Curcumin was administered
in the form of 500 mg capsules, and patients were advised to take the capsules after
meal. Administered curcumin had a phytosomal formulation (Meriva®; Indena S.p.A, Milan, Italy) that contained a complex of curcumin and soy phosphatidylcholine
in a 1:2 weight ratio, and 2 parts of microcrystalline to improve flowability, with
an overall content of curcumin in the final product of around 20% [47].
The study protocol was given approval by the institutional Ethics Committee and written
informed consent was obtained from participants. The clinical trial protocol has been
registered under the Iranian Registry of Clinical Trials (IRCT) ID: IRCT2015122525641N2.
Anthropometric measurements
Height and weight of subjects were measured with the accuracy of 0.1 cm and 0.1 kg,
respectively. Measurements were performed in the standing position, with minimal clothing
and no shoes at baseline and the 8th week. Body mass index was calculated as weight (kg) divided by height (m) squared.
Blood sampling and biochemical measurements
Fasted blood samples were collected from the drawn from a cubital vein at baseline
and after 8 weeks of supplementation. Blood samples were centrifuged for 10 min at
a speed of 2 000–1 500 rpm to separate the serum. Serum samples were kept at −80+°C
until analyses. Serum levels of alanine aminotransferase (ALT), aspartate aminotransferase
(AST), alkaline phosphatase (ALP), total and direct bilirubin were measured at baseline
and at the end of study using routine enzymatic assays with commercial kits (Pars
Azmoon, Tehran, Iran).
Liver doppler sonography
Sonography of the liver is performed when the patient has fasted for 8–12 h before
the examination. The examination is performed using 3.5-or 5-MHz convex probe.
Liver fat content and the severity of hepatic steatosis was evaluated at baseline
and after 8 weeks of supplementation using ultrasonography with a Mindray DC-8 diagnostic
ultrasound system (convex 3.5–5.0 MHz) at the beginning and the end of the study.
Ultrasound assessments were performed by the same expert radiologist blinded to the
type of allocation, and with the same instrument. Assessments were performed in the
fasted state (8–12 h), with the subjects in the supine position.
Right and left lobes of the upper and lower surfaces of liver were studied. Echogenicity
of the liver, the presence or absence of bulky tumors cystic or solid and calcification
was assessed. Intrahepatic bile ducts, portal vein and hepatic artery were evaluated.
Hepatic steatosis was graded as 0 (lack of fat accumulation), 1 (mild increase in
echogenicity with normal visualization of the diaphragm and intrahepatic vessel borders),
2 (moderate increase in echogenicity with slightly impaired visualization of the diaphragm
and intrahepatic vessel borders), and 3 (severe increase in echogenicity with markedly
impaired visualization of the diaphragm, intrahepatic vessel borders and the posterior
portion of the right hepatic lobe).
The portal vein was evaluated in a sagittal oblique view demonstrating the vessels’
longest axis. The antegrade hepatopetal flow throughout the entire cardiac cycle was
considered as the portal vein flow. The length of the liver on the sagittal view in
the mid-clavicular line was considered as liver size and was measured in the left
posterior-oblique position.
Statistical analysis
Statistical analyses were performed using the SPSS software version 11.5 (SPSS Inc.,
Chicago, Illinois, USA). The study population size was estimated at the significance
level of 95%, with a power of 80% and an effect size of 0.7 for AST and ALT. Data
were expressed as mean±SD or number (%). Within-group comparisons were performed using
paired samples t-test (for normally distributed data) or Wilcoxon signed-ranks test (for non-normally
distributed data). Between-group comparisons were performed using independent samples
t-test (for normally distributed data) or Mann-Whitney U test (for non-normally distributed
data). Categorical variables were compared using Fisher’s Exact test. Binary logistic
regression analysis was used to adjust for the effect of potential confounders on
the association between curcumin supplementation and changes the severity of NAFLD
according to ultrasonographic findings. For this analysis, change in the ultrasonographic
findings (categorized as either improvement or lack of improvement) was entered into
the model as dependent variable. A p-value of<0.05 was considered as statistically significant in all analyses. All analyses
were performed per protocol.
Results
Baseline characteristics
86 subjects completed the study. There were 6 drop-outs in the curcumin group and
9 in the placebo group. All drop-outs were because of stopping the medication owing
to the self-perception of lack of benefit ([Fig. 1]). Drop-out rate did not differ between the study groups (p=0.579). The study groups
were comparable in terms of age, gender, BMI, SBP, DBP, ALT, AST, ALP, total and direct
bilirubin, smoking history and frequency of type 2 diabetes, coronary heart disease,
obesity, hypertension, hyperlipidemia and hypertension ([Table 1]). However, there was a significant baseline difference in waist circumference between
the study groups ([Table 1]). The severity of fatty liver based on parenchymal echogenicity and hepatic vein
flow were comparable between the groups, yet portal vein diameter and liver volume
were different ([Table 1]).
Fig. 1 Flow diagram of study.
Table 1 Comparison of baseline and demographic characteristics between the study groups.
|
Curcumin
|
Placebo
|
p-Value
|
|
N
|
44
|
43
|
|
|
Female (%)
|
20 (45.5)
|
16 (37.2)
|
0.516
|
|
Age (y)
|
44.98±12.59
|
47.21±10.29
|
0.369
|
|
BMI (kg/m
2
)
|
28.97±3.42
|
29.07±3.48
|
0.899
|
|
Waist circumference (cm)
|
97.07±11.04
|
101.90±11.43
|
0.048
|
|
SBP (cmHg)
|
12.00±0.94
|
12.32±1.07
|
0.126
|
|
DBP (cmHg)
|
7.82±0.62
|
8.02±0.46
|
0.116
|
|
Total bilirubin (mg/dL)
|
0.73±0.17
|
0.77±0.23
|
0.354
|
|
Direct bilirubin (mg/dL)
|
0.42±0.15
|
0.36±0.14
|
0.075
|
|
ALT (U/L)
|
35.46±22.97
|
36.81±24.32
|
0.790
|
|
AST (U/L)
|
27.63±11.35
|
27.44±10.01
|
0.934
|
|
ALP (U/L)
|
172.52±40.65
|
170.77±40.80
|
0.841
|
|
Hepatic vein flow velocity (cm/sec)
|
17.15±2.16
|
17.29±3.44
|
0.829
|
|
Portal vein diameter (mm)
|
11.41±1.18
|
10.54±1.99
|
0.015
|
|
Liver size (mm)
|
154.75±11.62
|
160.12±11.89
|
0.036
|
|
Obesity (%)
|
12 (27.3)
|
14 (32.6)
|
0.644
|
|
Hypertension (%)
|
9 (20.5)
|
12 (27.9)
|
0.461
|
|
Dyslipidemia (%)
|
21 (47.7)
|
29 (67.4)
|
0.083
|
|
Diabetes (%)
|
11 (25.0)
|
10 (23.3)
|
1.000
|
|
CHD (%)
|
6 (13.6)
|
6 (14.0)
|
1.000
|
Values are expressed as mean±SD or number (%). SBP: systolic blood pressure; DBP:
diastolic blood pressure; ALT: alanine aminotransferase; AST: aspartate aminotransferase;
ALP: alkaline phosphatase; CHD: coronary heart disease
Anthropometrics and blood pressure
Supplementation with curcumin was associated with significant reductions in BMI (p<0.001)
and waist circumference (p<0.001) by the end of trial; however, no significant anthropometric
change was observed with placebo (p>0.05) ([Table 2]). Between-group comparison of changes revealed a significant difference between
curcumin and placebo groups with respect to BMI (p=0.003) and waist circumference
(p=0.024) ([Table 3]). With respect to blood pressure, there was no significant change in SBP and DBP
in any of the study group by the end of trial ([Table 2]), nor was there any significant difference between the groups in terms of blood
pressure changes during the course of study (p>0.05) ([Table 3]).
Table 2 Within-group comparison of biochemical parameters between curcumin and placebo
groups.
|
Curcumin
|
Placebo
|
|
Pre-trial
|
Post-trial
|
p-Value
|
Pre-trial
|
Post-trial
|
p-Value
|
|
BMI (kg/m2)
|
28.97±3.42
|
27.98±3.05
|
<0.001
|
29.07±3.48
|
28.92±3.69
|
0.461
|
|
Waist circumference (cm)
|
97.07±11.04
|
95.33±10.77
|
<0.001
|
101.90±11.43
|
101.66±11.25
|
0.664
|
|
SBP (cmHg)
|
12.00±0.94
|
12.07±0.69
|
0.417
|
12.32±1.07
|
12.23±0.75
|
0.493
|
|
DBP (cmHg)
|
7.82±0.62
|
7.89±0.62
|
0.439
|
8.02±0.46
|
7.91±0.65
|
0.251
|
|
Total bilirubin (mg/dL)
|
0.73±0.17
|
0.66±0.17
|
<0.001
|
0.77±0.23
|
0.72±0.18
|
0.001
|
|
Direct bilirubin (mg/dL)
|
0.42±0.15
|
0.40±0.14
|
0.162
|
0.36±0.14
|
0.37±0.14
|
0.793
|
|
ALT (U/L)
|
35.46±22.97
|
24.85±12.84
|
<0.001
|
36.81±24.32
|
41.33±23.97
|
<0.001
|
|
AST (U/L)
|
27.63±11.35
|
20.68±6.65
|
<0.001
|
27.44±10.01
|
31.23±12.80
|
<0.001
|
|
ALP (U/L)
|
172.52±40.65
|
172.91±39.85
|
0.912
|
170.77±40.80
|
172.59±43.37
|
0.399
|
|
Hepatic vein flow velocity (cm/sec)
|
17.15±2.16
|
21.35±3.35
|
<0.001
|
17.29±3.44
|
15.04±3.88
|
<0.001
|
|
Portal vein diameter (mm)
|
11.41±1.18
|
10.70±1.05
|
<0.001
|
10.54±1.99
|
11.49±1.61
|
<0.001
|
|
Liver size (mm)
|
154.75±11.62
|
152.82±10.17
|
0.003
|
160.12±11.89
|
161.88±27.03
|
<0.001
|
Values are expressed as mean±SD. SBP: systolic blood pressure; DBP: diastolic blood
pressure; ALT: alanine aminotransferase; AST: aspartate aminotransferase; ALP: alkaline
phosphatase
Table 3 Between-group comparison of biochemical parameters between curcumin and placebo
groups.
|
Curcumin
|
Placebo
|
p-Value
|
|
BMI (kg/m2)
|
−0.99±1.25
|
−0.15±1.31
|
0.003
|
|
Waist circumference (cm)
|
−1.74±2.58
|
−0.23±3.49
|
0.024
|
|
SBP (cmHg)
|
0.07±0.59
|
−0.08±0.96
|
0.185
|
|
DBP (cmHg)
|
0.07±0.59
|
−0.12±0.66
|
0.116
|
|
Total bilirubin (mg/dL)
|
−0.07±0.08
|
−0.06±0.11
|
0.271
|
|
Direct bilirubin (mg/dL)
|
−0.02±0.10
|
0.005±0.12
|
0.271
|
|
ALT (U/L)
|
−10.61±15.49
|
4.51±7.40
|
<0.001
|
|
AST (U/L)
|
−6.95±7.47
|
3.79±6.43
|
<0.001
|
|
ALP (U/L)
|
0.39±22.96
|
1.82±14.00
|
0.727
|
|
Hepatic vein flow velocity (cm/sec)
|
4.20±2.77
|
−2.25±2.78
|
<0.001
|
|
Portal vein diameter (mm)
|
−0.70±0.73
|
0.95±1.43
|
<0.001
|
|
Liver size (mm)
|
−1.93±4.00
|
1.76±22.89
|
<0.001
|
Values are expressed as mean±SD. SBP: systolic blood pressure; DBP: diastolic blood
pressure; ALT: alanine aminotransferase; AST: aspartate aminotransferase; ALP: alkaline
phosphatase
Liver doppler solography
Comparison of liver ultrasonographic findings revealed a significant improvement in
the curcumin vs. placebo group (p<0.001) ([Table 4]). Ultrasonographic findings were improved in 75.0% of subjects in the curcumin group,
while the rate of improvement in the placebo group was 4.7% (p<0.001). The frequency
of increased liver fat content in the curcumin group was 4.5% but 25.6% of subjects
in the placebo group had their liver fat content increased (p=0.007) ([Table 4]).
Table 4 Comparison of NAFLD severity within and between the study groups.
|
|
Curcuminoids
|
Placebo
|
p-Value*
|
|
Before
|
After
|
p
|
Before
|
After
|
P
|
|
|
NAFLD severity
|
Grade 0
|
0 (0)
|
15 (34.1)
|
<0.001
|
0 (0)
|
1 (2.3)
|
0.013
|
<0.001
|
|
Grade 1
|
17 (38.6)
|
21 (47.7)
|
17 (39.5)
|
10 (23.3)
|
|
Grade 2
|
23 (52.3)
|
6 (13.6)
|
19 (44.2)
|
21 (48.8)
|
|
Grade 3
|
4 (9.1)
|
2 (4.5)
|
7 (16.3)
|
11 (25.6)
|
* Comparison of change values between the study groups
Findings of Doppler solography indicated an increase in hepatic vein flow (p<0.001)
accompanied by reduction of portal vein diameter (p<0.001) and liver volume (p=0.001)
following treatment with curcumin, yet reverse effects were observed in the placebo
group ([Table 2]). Between-group comparisons revealed significant improvements in hepatic vein flow
(p<0.001), portal vein diameter (p<0.001) and liver volume (p<0.001) in the curcumin
vs. placebo group ([Table 3]).
Liver enzymes and bilirubin
Consistent with the findings of liver ultrasonography, serum levels of AST and ALT
were reduced by the end of trial in the curcumin group (p<0.001) but elevated in the
placebo group (p<0.001) ([Table 2]). The effect of curcumin in reducing serum AST (p<0.001) and ALT (p<0.001) levels
was also significant in the between-group comparison ([Table 3]). Comparison of baseline vs. end-trial serum ALP concentrations did not reveal a
significant alteration in any of the study groups ([Table 2]), nor was there a significant difference between the groups in terms of ALP changes
during the study period (p>0.05) ([Table 3]). With respect to total and direct bilirubin concentrations, no significant difference
was observed between curcumin and placebo groups (p>0.05) ([Table 2], [3]).
Regression analysis
Binary logistic regression analysis was performed to adjust the ultrasonographic findings
– as the primary efficacy measure – for baseline waist circumference, liver volume,
portal vein diameter, presence of dyslipidemia and serum direct bilirubin concentrations
as potential confounders of treatment response. Selection of confounders was based
on the baseline difference between curcumin and placebo groups using a p-value cut-off
of 0.1. The effect of curcumin in reducing liver fat remained significant (p<0.001)
after adjustment for the above-mentioned confounders ([Table 5]). The crude unadjusted effect of curcumin supplementation on the improvement of
liver fat content was 61.50 (95% CI: 12.73, 297.04) (p<0.001).
Table 5 Binary logistic regression analysis for the effect of curcumin supplementation
on liver fat content after adjustment for potential confounders.
|
Variables entered in the model
|
Odds ratio
|
95% CI
|
p-Value
|
|
Baseline waist circumference
|
1.02
|
0.96, 1.09
|
0.432
|
|
Baseline liver size
|
0.98
|
0.92, 1.03
|
0.420
|
|
Baseline portal vein diameter
|
0.78
|
0.52, 1.18
|
0.237
|
|
Curcumin treatment
|
96.50
|
14.30, 651.22
|
<0.001
|
Change in the ultrasonographic findings (categorized as either improvement or lack
of improvement) was entered into the model as dependent variable, and the variables
listed in the table were entered as independent factors
Safety
Curcumin was safe and well tolerated in this trial. There was no report of severe
adverse events.
Discussion
Findings of the present study suggest that supplementation with phytosomal curcumin
is well tolerated and significantly reduces fatty liver parameters in individuals
with NAFLD. This is consistent with the results of previous studies showing that curcumin
regulates hepatic lipogenesis through AMP-activated protein kinase activation which
leads to inhibition of hepatic lipid accumulation [48]. Thus, curcumin can blunt the “first hit” of NAFLD pathogenesis which involves development
of hepatic steatosis [9]. Taking both ultrasonographic and biochemical (serum transaminase changes) findings
together, it appears that the efficacy of curcumin supplementation in patients with
NAFLD is not only statistically significant but also clinically relevant owing to
the high improvement rate (75.0%) that was observed. However, a more definite judgment
on the clinical relevance of the effects necessitates longitudinal designs assessing
the efficacy of curcumin in preventing or slowing the progression of NAFLD into severe
outcomes such as NASH and cirrhosis.
Several studies have indicated potential benefits of curcumin in the prevention and
treatment of obesity, diabetes, and metabolic syndrome [49]. Curcumin decreases the production of reactive oxygen species by inhibition of hepatic
stellate cell activation and reduction of tissue inhibitor of metalloprotease-1 release
[50]. Another possible mechanism of action of curcumin may involve peroxisome proliferator-activated
receptor-γ (PPAR-γ) activation which can improve insulin sensitivity [51]. Moreover, the potent antioxidant effect of curcumin in lowering lipid peroxidation
and oxidative stress in different tissues including the liver has been previously
described [52]. These antioxidant effects of curcumin may be explained by its metal chelating ability
modulatory effect on the expression of antioxidant enzymes such as superoxide dismutase,
catalase and glutathione peroxidase [53]
[54].
Different histological changes may occur during NAFLD including macrovesicular steatosis,
lobular inflammatory infiltration, hepatocellular ballooning, formation of Mallory
bodies, and perisinusoidal fibrosis [55]
[56]
[57]
[58]. It has been demonstrated that curcumin supplementation reduces aminotransferases
levels and improves both steatosis and inflammation in liver tissues [59]. Moreover, curcumin inhibits NF-κB activation, decreases downstream induction of
ICAM-1, COX-2 and MCP-1, and reduces intrahepatic gene expression of monocyte chemoattractant
protein-1, CD11b, procollagen type I, and tissue inhibitor of metalloprotease-1 leading
to the mitigation of the development and progression of hepatic inflammation and fibrosis
[50]
[60]. Also, curcumin significantly decreases serum levels of TNF-alpha and IL-6 [61], 2 pro-inflammatory cytokines involved in the “second hit” of the pathogenesis of
NAFLD [9]. Lifestyle modification through adherence to a healthy diet and optimal physical
activity is the first step in the prevention and treatment of NAFLD [62]
[63]. In the present study, all patients received dietary and lifestyle advises. Along
with the observed improvements in the ultrasonographic and biochemical indices, curcumin
supplementation resulted in a significant reduction in anthropometric indices including
BMI and waist circumference. It is worth noting that curcumin has been shown to enhance
physical performance and reduce physiological fatique and [64] this might have contributed to the therapeutic effects of curcumin in reducing BMI
and other indices of NAFLD.
On the other hand, our results support the efficacy and safety of curcumin supplementation
as it has been previously reported by several studies in both healthy subjects and
patients with different diseases [65]
[66]
[67]
[68].
The main limitation of the present study was that histological changes were not assessed
because liver biopsy is an invasive diagnostic test limited by ethical issues; therefore,
hepatic findings were evaluated by hepatic ultrasound which has been shown to have
high sensitivity, specificity and accuracy for the assessment of NAFLD [69]
[70]. Besides, some other indices such as gamma-glutamyltransferase and spleen size that
could serve as indices of NAFLD progression were not assesse din this study and deserve
to be measured in future trials of curcumin. Another limitation is the short duration
of follow-up which impedes judgment on the long-term efficacy of curcumin supplementation
particularly in terms of reducing vascular and hepatic outcomes. Nonetheless, that
curcumin exerts protective effects as early as 8 weeks of supplementation is a proof
for the relevance of supplementation with this agent. Finally, given the small population
size, some between-group differences in baseline characteristics were present in spite
of randomization. Hence, confirmatory evidence from intention-to-treat analyses in
larger trials would be ideal to confirm the positive findings observed in the present
study.
In conclusion, results of this study revealed that short-term curcumin supplementation
improves sonographic findings of NAFLD and hepatic transaminase levels. However, future
studies are warranted to elucidate if the anti-steatotic properties of curcumin are
dose-dependent, and also the value of curcumin in reducing cardiovascular and hepatic
endpoints such as cirrhosis and hepatocellular carcinoma.
