Key words
zinc - supplementation - leptin - meta-analysis
Introduction
Obesity is a pathological condition characterized by an imbalance between receiving
and consuming energy. In 2016, World Health Organization reported that 650 million
adults throughout the world were obese [1]. It has been established that obesity is a major contributor to mortality in developed
countries [2]. Moreover, obesity is deemed as a risk factor for cardiovascular diseases, hypertension,
diabetes mellitus and some cancers [3]
[4]
[5].
Several biological agents such as leptin and zinc-α2-glycoprotein are implicated in body weight regulation and metabolic homeostasis [6]
[7]
[8]. For example, zinc-α2-glycoprotein is involved in the regulation of adipose tissue metabolism [9]
[10]. Leptin is a known adipokine that is implicated to have biological functions in
inflammation, reproduction, angiogenesis and bone formation [11]
[12]
[13]
[14]. Leptin, as an indicator of adipose tissue mass, affects the hypothalamus through
negative feedback, thus promotes satiety and subsequently reduces food intake [15]
[16]. Moreover, high serum leptin level is associated with renal impairment [17]. Recently, the association between serum leptin level and several other elements
such as zinc, has been extensively studied.
Zinc is a trace element and is required for the optimal activity of multitudinous
proteins including enzymes, gene expression regulatory proteins, receptors, and membrane
proteins. Thus, zinc participates in almost all the metabolic pathways [18]. Moreover, zinc affects insulin function and consequently carbohydrate metabolism
[19]
[20]. In some studies, low dietary intake and low serum level of zinc are associated
with higher incidence of diabetes, insulin resistance, cardiac diseases, hypertension,
and some cancers [21]
[22]. Unlike some other elements, zinc is generally not stored in body, therefore, it
should be received via food or supplement [23].
According to previous studies, zinc supplementation has a controversial effect on
serum leptin levels. Zinc is connected to insulin signaling and in some situations
like obesity and end-stage renal disease [24]
[25]
[26]
[27]
[28]. So far, zinc has been widely studied for its therapeutic and preventive features
and its association with serum leptin level is investigated in several studies. Some
of these trials found a direct association between zinc intake and leptin level [25]
[29]
[30]
[31], while others, among which a study conducted on Ache males of eastern Paraguay,
did not find such an association [26]
[32]
[33]. Serum leptin levels in patients with renal failure are higher than normal subjects
due to renal filtration defects, and this leads to decreased appetite and protein-energy
malnutrition in these patients. The study of Argani et al. showed that zinc supplementation
decrease serum leptin levels [25]. According to Payahoo et al., leptin plays a key role in regulating body weight
and fat mass by influencing appetite and fuel utilization, and zinc supplementation
seems to increase serum leptin levels in obese individuals [28].
Due to the conflicting results, we performed a systematic review and meta-analysis
of published randomized control trials (RCTs) to clarify the nature of the association
between zinc supplementation and serum leptin level.
Materials and Methods
Search strategy
SCOPUS (http://www.scopus.com), Medline (http://www.ncbi.nlm.nih.gov/PubMed), and
Google Scholar were searched to find all the relevant clinical trials up to April
2018. We used the following search terms in our search in the aforementioned databases:
(“zinc [MeSH]” OR “zinc” OR “zn”) AND (“leptin [MeSH]” OR leptin OR “leptin level”
OR “adipokines”). Additionally, we also performed a hand-search of the reference lists
of retrieved articles and previous reviews to include other potentially eligible trials.
Language restrictions were not applied. Cochrane handbook for systematic reviews of
interventions and Preferred Reporting Items for Systematic Reviews and Meta-Analyses
(PRISMA) guidelines were followed through all steps of the study [34]
[35].
Selection criteria
Human trials were included in the meta-analysis if they fulfilled the following inclusion
criteria: (I) were randomized clinical trials with either parallel or crossover designs
in human adults; (II) reported corresponding serum leptin levels before and after
intervention in each group; (III) compared oral zinc supplementation with the placebo
group. Studies were excluded if they: (I) experimented other agents along with zinc;
(II) were non-clinical trials; (III) did not provide sufficient information on leptin
levels before and after the trial in placebo and intervention groups.
Data extraction and quality assessment
First author’s last name, country of origin, publication date, study population, gender,
sample size, zinc dose, treatment duration and means ± SD of leptin for intervention,
and placebo groups before and after the trial were extracted from included studies.
A first reviewer (MKh) performed the data extraction and numerical calculations, results
were double-checked by a second reviewer (MZ). Discrepancies were resolved through
discussion with a third reviewer (SMM). In trials with crossover design, only data
from the first part of the study (before washout period) were considered for analysis.
The quality of eligible articles was assessed using the Jadad scale based on the description
of randomization, allocation concealment, blinding, drop-outs and the presentation
of an intention-to-treat analysis [36]; trials with 3 or more points and 2 or fewer points were considered as a “high”
and “low” quality studies, respectively.
Statistical analysis
The effect size of meta-analysis was calculated based on mean differences and their
corresponding standard deviations (SDs) of changes in leptin levels for both intervention
and control groups [37]. In studies that reported the standard error of means (SEM), SD was calculated via
multiplying SEM by the square root of the sample size: SD=SEM × √n. The statistical
and between-study heterogeneity was evaluated using the Cochran’s Q-test and the I²
index. Subgroup analysis was done to identify potential sources of heterogeneity;
subgroups were based on the dose, trial duration, baseline BMI, gender, and mean age
of participants. To investigate the influence of an individual study on the overall
weight mean differences, a sensitivity analysis was performed. Meta-regression test
using the unrestricted maximum likelihood method was used to assess the relation between
pooled effect size and zinc supplementation dose and duration of treatment. Funnel
plot and Egger’s weighted regression tests were done to examine publication bias [38]. All statistical analyses were carried out via STATA software, version 14.0 (Stata
Corporation, College Station, TX, USA). p-Values less than 0.05 were considered statistically
significant.
Results
Study selection
A total of 663 records including 219 from PubMed and 444 from the Scopus was identified
following the initial literature search. The process of study selection is shown in
[Fig. 1]. After removing 131 duplicate records, 532 articles were assessed based on title
and abstract. Subsequently, after the title and abstract screening, 29 articles were
retrieved for full-text assessment. Among these remaining records, 23 articles were
excluded for the following reasons: were not clinical trials (n=5), were animal studies
(n=1) or done on children (n=2), not reporting sufficient data for baseline and/or
final leptin levels (n=6), done in combination with other components (n=4), without
placebo group (n=3) and same population data set (n=2). Finally, 6 eligible RCTs with
7 treatment arms were included in this meta-analysis.
Fig. 1 PRISMA flow chart of study selection process.
Study characteristics
The demographic characteristics of eligible studies are outlined in [Table 1]. Overall 244 subjects were enrolled in these trials (67% women). The mean age of
the subjects ranged from 23–56.6 years old and the BMI was between 21.6 and 34.7 kg/m2. Among these trials, 2 studies were carried out exclusively on women [39]
[40], 2 on men [26]
[31], one on both genders [28], and one study provided separate data for men and women, thus this study was counted
as two studies [25]. The studies were published from 2003 to 2014; they were conducted in various countries
including Paraguay [26], Mexico [31], Brazil [39], South Korea [40], and Iran [25]
[28]. Included papers were randomized clinical trials, all with double-blinded design
except one [26]. Four studies enrolled obese subjects [28]
[31]
[39]
[40], one Ache males [26], and another one was done on hemodialysis patients [25]. Supplementation dose varied from 25–100 mg/d and intervention duration ranged from
10 days to 8 weeks.
Table 1 General characteristics of the included studies.
|
First author (year) [Ref.]
|
Location
|
Study design
|
Gender
|
Mean age (year)
|
Baseline BMI
|
Patient features
|
Sample size
|
Duration (week)
|
Dose (mg/d)
|
Baseline leptin levels (ng/ml)
|
Jaded score
|
|
Bribiescas et al. (2003) [26]
|
Paraguay
|
Randomized, clinical trial
|
Males
|
47.2
|
22.8
|
Ache males
|
14
|
1.5
|
50
|
1.34
|
2
|
|
Gomez-Garcia et al. (2006) [31]
|
Mexico
|
Randomized, double-blinded, placebo-controlled
|
Males
|
25.5
|
30.7
|
Obese males
|
14
|
4
|
25
|
16.4
|
3
|
|
Marreiro et al. (2006) [39]
|
Brazil
|
Randomized, double-blinded, placebo-controlled
|
Females
|
35.5
|
35.8
|
Obese Women
|
56
|
4
|
30
|
23.6
|
3
|
|
Kim et al. (2012) [40]
|
South Korea
|
Randomized, double-blinded, placebo-controlled
|
Females
|
23
|
28.25
|
Obese Women
|
40
|
8
|
30
|
19.22
|
4
|
|
Argani et al. (2014) [25]
|
Iran
|
Randomized, double-blinded, placebo-controlled
|
Males
|
55.6
|
22.5
|
Hemodialysis patients
|
36
|
8
|
100
|
6.4
|
3
|
|
Argani et al. (2014) [25]
|
Iran
|
Randomized, double-blinded, placebo-controlled
|
Females
|
55.6
|
21.6
|
Hemodialysis patients
|
24
|
8
|
100
|
9
|
3
|
|
Payahoo et al. (2014) [28]
|
Iran
|
Randomized, double-blinded, placebo-controlled
|
Both
|
31.8
|
34.7
|
Obese subjects
|
60
|
4
|
30
|
35.8
|
3
|
The effect of zinc supplementation on serum leptin
Forest plot of the effectiveness of zinc supplementation on serum leptin is demonstrated
in [Fig. 2]. The pooled estimate from the random-effect model which was performed on 7 studies
with 121 cases and 123 controls, did not imply a significant change in serum leptin
levels after zinc supplementation (WMD: 0.74 ng/ml; 95% CI: −1.39 to 2.87, p=0.49).
The effect size was unaltered after sensitivity analysis. Also, a significant between-study
heterogeneity was found among studies (I2=81.7%, p=<0.001).
Fig. 2 Forest plot detailing weighted mean difference and 95% confidence intervals for the
impact of supplementation with zinc on serum leptin levels.
Subgroup analysis
Considering that the supplementation dose, intervention duration, participants’ gender,
body mass index (BMI), and age may influence the net changes of leptin, we performed
subgroup analysis on the basis of these variables to identify heterogeneity sources.
Findings from subgroup analysis are displayed in [Table 2]. The subgroup analysis showed that the dose of zinc supplementation (≥50 mg/d: I2=0.0%, p=0.75), intervention duration (≥ 6 weeks: I2=43.8%, p=0.16), participants’ gender (males: I2=28.4%, p=0.24), mean age of subjects (≥45 years: I2=0.0%, p=0.75), and baseline BMI (<25 kg/m2: I2=0.0%, p=0.75) were the potential sources of heterogeneity. Among these, a treatment
duration of greater than 6 weeks (WMD-1.71 ng/ml; 95% CI:−3.07 to −0.35, p=0.01) significantly
reduced leptin concentration compared with shorter duration (WMD: −0.16 ng/ml; 95%
CI:−0.47 to 0.13, p=0.27). In patients with female gender, also a significant reduction
in leptin was observed (WMD:−1.93 ng/ml; 95% CI: −3.72 to −0.14, p=0.03).
Table 2 Pooled estimates of effects on leptin within different subgroups.
|
Group
|
No of comparisons
|
WMD (95% CI)
|
p-Value
|
p-Heterogeneity
|
I2 (%)
|
|
Total
|
7
|
0.737 (−1.395, 2.869)
|
0.498
|
<0.001
|
81.7
|
|
Zinc dosage (mg)
|
|
|
|
|
|
|
<50
|
4
|
0.102 (−1.779, 1.983)
|
0.915
|
<0.001
|
90.7
|
|
≥50
|
3
|
−0.251 (−0.551, 0.049)
|
0.101
|
0.757
|
0.0
|
|
Intervention duration (week)
|
|
|
|
|
|
|
<6
|
4
|
−0.169 (−0.472, 0.134)
|
0.275
|
<0.001
|
87.8
|
|
≥6
|
3
|
−1.718 (−3.079, −0.356)
|
0.013
|
0.169
|
43.8
|
|
Gender
|
|
|
|
|
|
|
Male
|
3
|
−0.228 (−0.529, 0.072)
|
0.137
|
0.247
|
28.4
|
|
Female
|
3
|
−1.934 (−3.723, −0.145)
|
0.034
|
0.066
|
63.2
|
|
Mean age
|
|
|
|
|
|
|
<45
|
4
|
0.102 (−1.779, 1.983)
|
0.915
|
<0.001
|
90.7
|
|
≥45
|
3
|
−0.251 (−0.551, 0.049)
|
0.101
|
0.757
|
0.0
|
|
Baseline BMI
|
|
|
|
|
|
|
<25
|
3
|
−0.251 (−0.551, 0.049)
|
0.101
|
0.757
|
0.0
|
|
≥25
|
4
|
0.102 (−1.779, 1.983)
|
0.915
|
<0.001
|
90.7
|
Meta-regression
Meta-regression analysis was used to investigate the relationship between changes
in serum leptin levels and potential moderator variables. The results revealed that
the pooled estimate is independent of zinc dose (slope: − 0.0601; 95% CI: − 0.23,
0.11; p=0.41) and treatment duration (slope: −0.8642; 95% CI: −2.99, 1.26; p=0.34;
[Fig. 3]).
Fig. 3 Meta-regression plots of the association of mean changes in serum leptin levels with
dose and duration of zinc supplementation.
Publication bias
Visual inspection of funnel plot was not indicative of a significant publication bias
in this meta-analysis ([Fig. 4]). This observation was also upheld by the Egger’s linear regression (p=0.53).
Fig. 4 Funnel plot displaying publication bias in the studies reporting the impact of zinc
on serum leptin levels.
Discussion
The present meta-analysis was conducted on 6 eligible studies (with 7 treatment arms)
to determine the direction and magnitude of zinc’s influence on serum leptin. To the
best of our knowledge, this is the first meta-analysis of RCTs on this topic. Results
showed no significant association between oral zinc supplementation and serum leptin
level. Nevertheless, in subgroup analyses based on intervention duration and gender,
there was a statistically significant reduction in serum leptin in studies lasted
for more than 6 weeks and carried out on women.
Nowadays, obesity has become one of the most challenging issues in health-care field.
Complications of obesity such as hypertension, hyperlipidemia, diabetes, etc. are
themselves risk factors for many conditions and thus may lead to substantial mortality
[41]. Leptin serves as a biomarker of overweight and obesity and is secreted from adipose
tissue. At the cellular level, it seems to regulate food intake and energy expenditure
via different signaling pathways [42]. Moreover, leptin affects the function of other organs in pulmonary, circulatory,
reproductive, and nervous systems [43]. In obese individuals, leptin is elevated as a result of low sensitivity to circulating
leptin. This low sensitivity might account for the decreased efficacy of leptin in
suppressing food intake in these patients [44]. It has been suggested that leptin and insulin are biologically linked and affect
each other and they both are implicated in the pathophysiology of obesity [45]. As a trace element, Zinc serves several important functions in the body; its role
has been established in metabolism, appetite, taste, skin and hair health, the function
of various enzymes and above all, insulin performance [46]
[47]
[48]. It has been shown that zinc impacts insulin’s production, storage, and release
and seems to decrease insulin sensitivity [49]. Accordingly, an association between zinc itself and leptin might exist which may
consequently affect obesity.
In vitro and in vivo studies indicated that zinc deficiency seems to decrease expression
of leptin gene. In an animal study, zinc deficiency caused a decrease in leptin gene
expression, synthesis, and secretion corresponding to adipose tissue mass in rats
[50]. Also, Kwun et al. showed that mRNA of leptin significantly decreased following
zinc deficiency in rats [51].
Argani et al. showed that 100 mg daily oral zinc supplementation for 8 weeks significantly
decreased mean leptin level in hemodialytic women [25], however, in another study, daily supplementation with 30 mg of zinc for 4 weeks
did not change leptin levels significantly in obese women [39]. Moreover, zinc supplementation did not change leptin levels in Ache males [26]. On the contrary, Payahoo et al. showed that daily supplementation with 30 mg of
zinc significantly increased leptin level in obese subjects [28].
According to the results of subgroup analysis based on gender, zinc supplementation
significantly decreased leptin levels in women. Physiologically, leptin levels are
higher in women than in males due to the higher fat mass in women. Considering that
women had a higher baseline of leptin [52], the amount of reduction in leptin was statistically significant in comparison to
that in men. Also, in studies that lasted for more than 6 weeks, zinc supplementation
led to a significant decrease in leptin level. Zinc reduces leptin through gene expression
of involved enzymes or proteins, this possibly explains the significance of results
with longer duration. These results also indicate that zinc supplementation may be
beneficial for patients with renal insufficiency, because of the impairment in glomerular
filtration, the level of leptin in these patients increases, which can reduce appetite
and subsequently increase protein-energy malnutrition [53].
The current study has some limitations that should be noted. The included RCTs possessed
relatively small sample sizes (only one study had 60 patients), which resulted in
poor statistical power to detect meaningful effects in individual trials and in the
overall analysis. In addition, the eligible trials were heterogeneous due to the sample
size, dose, and duration of intervention, gender and age of participants. Also, various
chemical forms of zinc have been used in included studies including gluconate, acetate,
and sulfate, which have different oral absorption, and bioavailability. According
to the WHO, zinc sulfate (23% zinc), zinc acetate (30% zinc), and zinc gluconate (14%
zinc) have similar bioavailibilities. [54].
Conclusion
Results from this meta-analysis did not support the notion that zinc supplementation
may tend to decrease serum leptin level. However, zinc supplementation may significantly
reduce leptin in females and in interventions with duration of more than 6 weeks.
Additional studies with larger sample sizes, different doses, and duration should
be performed to approve our findings.
Author Contributions
MKh designed and SMM supervised the study. MZ, AS, and JR conducted literature searches,
data extraction, and independent search and reviewing. AT, NA, and HKV performed the
statistical analysis and composed the initial draft of the manuscript. All authors
read and approved the final version of the manuscript.
