Efficacy of omega-3 polyunsaturated fatty acids supplementation in managing overweight and obesity: A meta-analysis of randomized clinical trials

Abstract

Objective

Studies in rodents and humans have indicated that omega-3 polyunsaturated fatty acids (n-3 PUFA) may reduce weight. The aim of this meta-analysis was to evaluate evidence for the efficacy of n-3 PUFA in managing overweight and obesity.

Methods

We performed a systematic search of PubMed, Embase, and Cochrane Central Register of Controlled Trials until May 2015. Two reviewers independently determined the eligibility of studies and assessed the reporting quality of included randomized controlled trials (RCTs).

Results

A total of 11 RCTs involving 617 participants were included in this meta-analysis. Based on the meta-analysis of nine studies, a statistically nonsignificant difference was revealed in weight loss between n-3 PUFA and placebo (p=0.99; weighted mean difference [WMD]: 0.00; 95% confidence interval [CI] −0.42 to 0.43), whereas n-3 PUFA was superior to placebo in reducing serum triglyceride levels (p=0.0007; standard median difference [Std MD]: −0.59; 95% CI −0.93 to −0.25). Based on meta-analysis of seven studies, the analysis of aggregated data showed a significant reduction in waist circumference (p=0.005; WMD: −0.53; 95% CI −0.90 to −0.16). There were no significant differences in body mass index, total serum levels of cholesterol, low density lipoprotein cholesterol, high density lipoprotein cholesterol, and fasting glucose levels.

Conclusions

The evidence from RCTs showed that n-3 PUFA might effectively reduce waist circumference and triglyceride levels in overweight and obese adults, but n-3 PUFA may not effectively reduce body weight. Given the small number and poor quality of RCTs included in the meta-analysis, these results are inconclusive. A large-scale, well-designed RCT is needed to further address this issue..

Introduction

In only the past few decades, the dramatic prevalence of the obesity epidemic has become a public health issue of worldwide importance (1). This is partly due to the fact that obesity is associated with an increased risk of dyslipidemia, hypertension, diabetes, and cardiovascular disease (CVD). Therapeutic options for the treatment of obesity include dietary management (2), drug therapy, and bariatric surgery (3). Despite the wide range of treatments, dietary intervention is the cornerstone of managing obesity and related morbidities (4). Numerous dietary supplements are being marketed as slimming aids. The efficacy of these food supplements has not been proven, yet they are sold as over-the-counter preparations, and on the internet. One such supplement is omega-3 polyunsaturated fatty acids (n-3 PUFA).

Humans acquire n-3 PUFA from fish oils, which are obtained from the human diet by eating oily fish, such as herring, mackerel, salmon, albacore tuna, and sardines, or by consuming fish oil supplements or cod liver oil (5, 6). Marine microorganisms are the original source of the n-3 PUFA found in fish oils. There is substantial evidence that the consumption of n-3 PUFA reduces the risk of CVDs (5). In accordance with these studies, the American and European Societies of Cardiology have recommended that all patients with CVD consume 1 g of n-3 PUFA daily (5, 7).

Most of the evidence of the benefits of n-3 PUFA has been obtained for long-chain fatty acids, eicosapentaenoic acid (EPA), and docosahexaenoic acid (DHA), the active components of fish oils (8). However, there is no consensus on the beneficial effects of n-3 PUFA supplementation on reducing body weight. Some studies with rodents suggest that supplementation with n-3 PUFA may reduce weight or fat mass or protect against weight gain (9, 10), whereas other studies have shown no significant effects (11, 12). Similarly, there is growing evidence that n-3 PUFA can reduce body fat in humans by suppressing the appetite, and increasing fat oxidation and energy expenditure; however, there is still much controversy about these data, which have made it difficult to draw any definitive conclusions (13). Therefore, we conducted a meta-analysis to quantitatively summarize and critically evaluate the evidence from randomized clinical trials (RCTs) involving the use of n-3 PUFA as a weight loss supplement.

Methods

Search strategy

The electronic databases of PubMed, Embase, and Cochrane Central Register of Controlled Trials were searched (updated to May 20, 2015) using the following search terms: “overweight” or “obesity” or “weight loss” or “body weight” or “body fat” or “adiposity” or “body mass index (BMI)” and “linolenic acid” or “timnodonic acid” or “alpha-linolenic acid” or “ALA” or “eicosapentaenoic acid” or “EPA” or “docosahexaenoic acid” or “DHA” or “omega-3 fatty acid” or “n-3 fatty acid” or “fish oil”. All of the indexed studies were retrieved, and the reference lists of the identified publications were reviewed for additional pertinent studies. Non-English literature was not included. The literature search was carried out independently by two investigators.

Inclusion criteria

Trials that met the following criteria were included: (1) randomized blind placebo-controlled trials; (2) overweight (BMI ≥25 to 29.9 kg/m2) or obese (BMI ≥30 kg/m2) patients; (3) no history of other diseases like diabetes, coronary heart disease, stroke, cancer, hepatic disorder, a chronic inflammatory condition, or psychiatric disorders; (4) the patients had used n-3 PUFA supplements at least 1 gram/day dosage; (5) studies were included irrespective of whether or not they incorporated lifestyle changes into their trial regimen; (6) studies involving the use of n-3 PUFA as part of a combination product or treatment package were excluded from the systematic review; (7) primary outcomes reported included at least a body composition and/or lipid profile.

Data extraction and quality assessment

One reviewer screened the titles and abstracts of the RCTs that we identified. Full text articles were obtained for those trials that fulfilled the inclusion criteria or for which sufficient information was given. Two reviewers independently extracted data from trials that met the inclusion criteria on an Excel spreadsheet; any discrepancies in extracted data were resolved by a group discussion and consensus and final arbitration by the Cochrane editorial base. Attempts were made to seek further information from the authors of the original studies if data were unclear or incomplete. The methodologic quality criteria of randomization, allocation concealed, blinding, and intention-to-treat (ITT) analyses were graded as adequate, inadequate, and unclear. Studies were categorized as double blinding, single blinding, or unclear.

Statistical analysis

In our meta-analysis, body composition such as body weight, BMI, and waist circumference were considered primary outcomes. Because obesity was strongly associated with dyslipidemia and may account for the increased risk of type 2 diabetes, other secondary outcomes included lipid profile and fasting glucose, such as triglyceride, total cholesterol (TC), low density lipoprotein-cholesterol (LDL-C), and high density lipoprotein-cholesterol (HDL-C). All outcomes extracted from the literature were continuous data.

Review Manager software 5.2 provided by the Cochrane Collaboration was used for the meta-analysis. For continuous data, weighted mean differences (WMDs) or standardized mean differences (Std MDs) with their 95% confidence intervals (95% CIs) were calculated. Analyses were separately performed for each outcome. Heterogeneity across studies was assessed by the Cochrane's Q-test. The fixed-effects model was used to calculate the total effect size where Cochrane's Q-test p>0.10 and the I2 statistic I2<50% indicated statistical homogeneity. If heterogeneity of p<0.10 or I2>50% was found among the trials, a random-effects model was chosen. Two-tailed p-values ≤0.05 or 95% CIs not containing 0 (WMD) were considered statistically significant.

Results

Eligible studies and baseline characteristics

We identified 320 articles from our initial electronic search, and of these, 11 RCTs (14., 15., 16., 17., 18., 19., 20., 21., 22., 23., 24.) fulfilled the criteria for inclusion, and were selected for meta-analysis after application of the search criteria and quality assessment (Figure 1). In total, the studies included 617 participants. The key details of each study included in this systematic review are summarized in Table 1.

Figure 1.

Flow diagram of the study selection process for the meta-analysis. RCT: randomized controlled trial

Table 1.

Characteristics of studies included in the meta-analysis

First author Year Country	Daily dosage (Formulation)	Randomized/ Analyzed	Gender (male/ female)	Treatment duration	Adverse events	Control for lifestyle factors	Randomization appropriate	Allocation concealed	Sample size determined	Groups similar at baseline	Blinding	ITT analysis
Chan 2003(14) Australia	4 g(capsules)	20/20	20/0	6 weeks	None	Isocaloric diets and maintain their accusto- med degree of exercise	Unclear	Unclear	Unclear	Yes	Double blinding	Unclear
Krebs 2006(15) UK	5 g(capsules)	116/93	0/93	24 weeks	Not reported	800–900 kcals/day (12 weeks), 2500 kcal/ day(followed 12 weeks), exercise	Unclear	Unclear	Unclear	Yes	Double blinding	Unclear
Kunesova 2006(16) Czech Republic	2.8 g(capsules)	20/20	0/20	3 weeks	Not reported	2200kJ/day, exercise 60 min/day	Unclear	Unclear	Unclear	Yes	Unclear	Unclear
Thorsdottir 2007(17) Iceland	1.5 g(capsules)	160/134	52/82	8 weeks	Not reported	1350kcal/ day(males), 1579kcal/ day(females)	Unclear	Unclear	Unclear	Yes	Single blinding	Unclear
Hill 2007(18) Australia	6 g(capsules)	38/35	13/22	12 weeks	Not reported	Normal diet, normal level of physical activity.	Unclear	Unclear	Unclear	Yes	Double blinding	Unclear
Defina 2011(19) USA	3 g(capsules)	128/128	40/88	24 weeks	Gas- trointes- tinal discomfort	Calorie- controlled diet, exercise	Unclear	Unclear	Unclear	Yes	Single blinding	Adequate
Gam- melmark 2012(20) Denmark	2 g(capsules)	50/50	24/26	6 weeks	Gas- trointes- tinal discomfort	Normal diet	Unclear	Unclear	Unclear	Yes	Double blinding	Unclear
Munro 2012(21) Australia	6 g(capsules)	40/32	6/26	14 weeks	Not reported	3000kJ/day(4 weeks), healthy diet (followed 10 weeks)	Adequate	Adequate	Unclear	Yes	Double blinding	Unclear
Munro 2013(22) Australia	6 g(capsules)	42/39	9/30	8 weeks	Not reported	healthy diet(4 weeks), 3000kJ/day (followed 4 weeks)	Adequate	Adequate	Unclear	Yes	Double blinding	Unclear
Munro 2013(23) Australia	6 g(capsules)	43/33	11/22	12 weeks	Not reported	6000kJ/ day(males), 5000kJ/ day(females)	Adequate	Unclear	Unclear	Yes	Double blinding	Unclear
Spencer 2013(24) USA	4 g(tablets)	33/33	11/22	12 weeks	Not reported	No special diet regimen prescribed	Unclear	Unclear	Unclear	Yes	Unclear	Unclear

ITT: intention -to -treat.

Most of the included RCTs had flaws in the reporting of their methodology. Only three studies reported adequate randomization (21., 22., 23.), and two of the studies reported adequate allocation concealment (21, 22). Seven of eleven studies were double-blinded (14., 15., 18, 20., 21., 22., 23.). However, all included studies either did not mention or fully explain ITT analysis, except one study that used ITT analysis (19).

Most of RCTs included in this review incorporated lifestyle adjustments into their trial regimen. The daily caloric intake of participants ranged from 528 kcal to more than 2500 kcal. Participants in three RCTs (18, 20, 24) were allowed to continue their normal lifestyle, whereas the remaining participants had their daily caloric intakes monitored. Participants in one RCT were fed a normocaloric diet (14). In three studies, the lifestyle adjustments were divided into two phases; namely, weight-loss and weight-maintenance phases, in which the participants were required to consume different caloric diets (15, 21, 22). The dosage of n-3 PUFA used in the included RCTs ranged from 2 to 6 g. Treatment duration ranged from 3 to 24 weeks.

Meta-analyses of primary outcomes

The changes in body weight were evaluated in the nine studies (15., 16., 17., 18., 19., 21., 22., 23., 24.), which included 521 cases. The meta-analysis revealed a statistically nonsignificant difference in weight loss between n-3 PUFA and placebo (p=0.99; WMD: 0.00; 95% CI −0.42 to 0.43), without evidence of heterogeneity (p=0.09, I2=42%). Seven studies compared the effects of n-3 PUFA and placebo on waist circumference (15., 16., 17., 19, 21., 22., 23.). Analysis of aggregated data showed a significant reduction in waist circumference (p=0.005; WMD: −0.53; 95% CI −0.90 to −0.16), without evidence of heterogeneity (p=0.47, I2=0%). The changes in BMI were evaluated in the nine studies (15., 16., 17., 18., 19., 20., 21., 22., 23.). Analysis of aggregated data revealed no significant difference associated with the administration of n-3 PUFA (p=0.18; WMD: −0.19; 95% CI −0.46 to 0.09), although heterogeneity existed among these studies (p=0.01, I2 =60%; Figure 2).

Figure 2.

Forest plot of studies that evaluated the effect of n-3 PUFA on primary outcomes compared with placebo. Each block represents a study. Size of square is proportional to the precision of the estimate. Each square represents the standardized mean difference (SMD) for each study with 95% confidence interval (CI) indicated by horizontal line. BMI: body mass index

Meta-analyses of secondary outcomes

Nine studies (14, 15, 18., 19., 20., 21., 22., 23., 24.), which included 441 cases, compared the effects of n-3 PUFA and placebo on triglyceride levels. However, significant heterogeneity was observed among these studies (p=0.005, I2=64%). The random effects model of meta-analysis was used to combine the effect size. n-3 PUFA was superior to placebo in reducing serum triglyceride levels (p=0.0007; Std MD: −0.59; 95% CI −0.93 to −0.25). Eight studies compared the effects of n-3 PUFA and placebo on TC levels (14, 15, 18, 20., 21., 22., 23., 24.). There were no significant changes in the concentration of TC (p=0.19; Std MD: −0.29; 95% CI −0.74 to 0.15), with heterogeneity existing among these studies (p=0.0007, I2=72%). Nine studies included data about HDL-C (14, 15, 18., 19., 20., 21., 22., 23., 24.) and eight studies included data about LDL-C (14, 15, 19., 20., 21., 22., 23., 24.). The meta-analysis showed that the effects of n-3 PUFA on HDL-C and LDL-C levels did not significantly differ from placebo (p=0.29; Std MD: 0.16; 95% CI −0.14 to 0.46 and p=0.80; Std MD: 0.11; 95% CI −0.77 to 1.00, respectively). High heterogeneity was detected with HDL-C and LDL-C variables (p=0.02, I2=56% and p<0.00001, I2=94% respectively). The changes in fasting glucose levels were evaluated in seven studies (14, 15, 19., 20., 21., 22., 23.). After analysis of the aggregated results, we found that there was no significant difference associated with administration of n-3 PUFA (p=0.50; WMD: −0.15; 95% CI −0.60 to 0.29), although heterogeneity existed among these studies (p=0.0004, I2=76%; Figure 3).

Figure 3.

Forest plot of studies that evaluated the effect of n-3 PUFA on secondary outcomes compared with placebo. See Figure 2 for the legend of symbols used. TC: total cholesterol; HDL-C: high density lipoprotein-cholesterol; LDL-C: low density lipoprotein-cholesterol

Discussion

This study is the first meta-analysis to analyze the effects of n-3 PUFA on managing overweight and obesity in placebo-controlled RCTs. Although significant differences between groups were found for some parameters, the limited evidence suggests that n-3 PUFA intake does not generate a significant decrease in body weight when compared to placebo. In addition, the meta-analysis results revealed a statistically significant difference in loss of waist circumference and triglyceride levels favoring n-3 PUFA over placebo, although there was no significant difference in BMI, serum levels of TC, HDL-C, LDL-C, and fasting glucose.

n-3 PUFA is postulated to exert its weight loss effect by increasing body fat oxidation and energy expenditure, changing adipocyte apoptosis, improving circulation that might facilitate nutrient delivery to skeletal muscle, and changing gene expression to shift metabolism towards increased accretion of lean tissue that suppressed fat deposition. It is claimed that n-3 PUFA may also reduce obesity that accompanies appetitesuppressing effects (13, 25). The reasons for this are speculated to be associated with reduced abdominal obesity as a result of a reduction in body fat. Hill et al. (18) investigated the effect of n-3 PUFA supplements and regular exercise on body composition in overweight or obese adults, and found that n-3 PUFA supplementation reduced body fat, even when BMI and body weight were unchanged.

Although some reviews have recently been published about the anti-obesity effects of n-3 PUFA (13, 25), the quality of these reviews was limited, and they did not present specific methods on data extraction or quality assessment. The following factors have strengthened this meta-analysis. First, studies were included or excluded according to strict criteria. Next, the meta-analysis offered an up-to-date and complete overview of all RCTs involving the efficacy of n-3 fatty acid supplementation in managing overweight and obesity, because it was the result of an extensive search, including gray literature and unpublished studies. Finally, for each outcome, the unit was converted to an international system of units.

There are several potential limitations that should be considered when interpreting the results of this meta-analysis. First, not all studies included data about body weight, waist circumference, BMI, lipid profile and glucose. Our analysis was based on a small number of RCTs and RCTs of poor quality, thereby limiting the reliable results. Second, there were methodological issues related to study design. Allocation concealment was clearly described in only two studies (21, 22); hence, selective bias may have existed in the trials. The withdrawal rates of the RCTs ranged from 0% to 23.25%, which may have affected the accuracy of the results. ITT analysis was performed in only one study with patient withdrawals (19). Attrition bias cannot be ignored in studies with withdrawals where ITT analysis was not performed. Next, our findings were limited to Western populations because of a paucity of data in Eastern populations. Finally, significant heterogeneity was observed for a number of pooled variables when analyzing the outcomes for our current meta-analysis. Such heterogeneity confounds interpretation of statistical findings. We had initially planned to conduct subgroup analyses; for example, according to the type of experimental design and age of selected patients, as well as drug dose, types of n-3 PUFA supplementation, duration of treatment, and units of laboratory tests, all of which differed between studies. However, there was not a sufficient number of trials to perform this analysis. Therefore, the random-effects model was adopted, although it cannot completely eliminate heterogeneity.

This meta-analysis provides additional information that may be useful for future studies on the topic. First, adipose tissue is an endocrine organ that secretes several active substances, including a range of inflammatory mediators known as adipokines and cytokines. Some studies suggest that insulin resistance increases with intra-abdominal adipose tissue size in obese individuals. 7 of 11 studies described the fat mass of subjects in the n-3 PUFA group in detail (15, 17., 18., 19., 21., 22., 23.), which applied different kind of techniques such as dual X-ray absorptiometry and bioelectrical impedance to assess. Included studies to conduct the analysis of fat mass which used uniformly standard measurement could make the outcome more convincing. Future studies may focus on the effect of n-3 PUFA on fat mass and inflammatory markers in overweight and obesity subjects, such as tumor necrosis factor-a and interleukin-6, C-reactive protein and leptin, because this might be a potential action point for prevention of CVD (26). Second, lifestyle factors such as diet and physical exercise are a very important aspect of weight control (27). However, there was considerable difference in the average daily caloric intake and level of physical activity undertaken by study participants. Most studies lacked objective outcome measures to estimate the extent to which these variations influenced the outcome of the study result. Whether a multidomain intervention consisting of nutritional counseling and physical exercise in combination with n-3 PUFA supplementation is more beneficial than a single intervention is unclear. Therefore, focusing on these supplementary clues may be useful in future research studies on the topic (28). Third, the most commonly observed adverse effects of n-3 PUFA supplementation are nausea, gastrointestinal upset, and a “fishy” burp (8). However, only 3 of 11 studies described the adverse reactions of subjects in the n-3 PUFA group in detail (14, 19, 20). The follow-up period ranged from 3 to 24 weeks. Considering the fact that these studies were of short duration, the safety of long-term n-3 PUFA intake seems is unclear. Thus, it is essential for investigators of future trials to incorporate surveillance time frames into the clinical trials to monitor any medium- and long-term adverse events associated with the use of n-3 PUFA (29). Finally, a gender difference was observed in two studies (17, 22), which could also be interpreted as an additional benefit of the study, since this pronounced gender difference was unexpected. This may partly be explained by the fact that women naturally lose less weight and are more responsive to n-3 PUFA metabolism than men (30). In addition, a possible reason for the different outcomes from these studies could be the different amounts of n-3 PUFA supplementation and the relative proportions of EPA to DHA. Future studies should focus on the effects of gender, n-3 PUFA formulation, dosage, and proportion of EPA to DHA.

Conclusions

The current limited evidence from RCTs suggests that n-3 PUFA supplementation might effectively reduce waist circumference and triglyceride levels, but n-3 PUFA may not effectively reduce body weight. In addition, n-3 PUFA supplementation had no significant effects on BMI or serum levels of TC, HDL-C, LDL-C, and fasting glucose. Large-scale, multi-center, and placebo-controlled long-term trials should be rigorously designed to substantiate the current findings and to investigate the long-term effects of n-3 fatty acid supplementation in managing overweight and obesity.

Conflict of Interest/Financial Disclosures:

None reported.