복부비만

Despite decades of unequivocal evidence that waist circumference provides both independent and additive information to BMI for predicting morbidity and risk of death, this measurement is not routinely obtained in clinical practice. This Consensus Statement proposes that measurements of waist circumference afford practitioners with an important opportunity to improve the management and health of patients. We argue that BMI alone is not sufficient to properly assess or manage the cardiometabolic risk associated with increased adiposity in adults and provide a thorough review of the evidence that will empower health practitioners and professional societies to routinely include waist circumference in the evaluation and management of patients with overweight or obesity. We recommend that decreases in waist circumference are a critically important treatment target for reducing adverse health risks for both men and women. Moreover, we describe evidence that clinically relevant reductions in waist circumference can be achieved by routine, moderate-intensity exercise and/or dietary interventions. We identify gaps in the knowledge, including the refinement of waist circumference threshold values for a given BMI category, to optimize obesity risk stratification across age, sex and ethnicity. We recommend that health professionals are trained to properly perform this simple measurement and consider it as an important ‘vital sign’ in clinical practice.

Excessive visceral adipose tissue appears to trigger a cascade of metabolic disturbances that seem to coexist with ectopic fat storage in muscle, liver, heart and the ß-cell. Therefore, the reduction of visceral adipose tissue potentially plays a pivotal role in the treatment of the metabolic syndrome. The purpose of this systematic review and meta-analysis is to describe the overall effect of exercise on visceral adipose tissue and to provide an overview of the effect of different exercise regimes, without caloric restriction, on visceral adipose tissue in obese persons. A systematic literature search was performed according to the PRISMA statement for reporting systematic reviews and meta-analyses. The initial search resulted in 87 articles after removing duplicates. After screening on title, abstract and full-text 15 articles (totalling 852 subjects) fulfilled the a priori inclusion criteria. The quality of each eligible study was assessed in duplicate with “The Critical Review Form for Quantitative Studies”. Using random-effects weights, the standardized mean difference (Hedge's g) of the change in visceral adipose tissue was −0.497 with a 95% confidence interval of −0.655 to −0.340. The Z-value was −6.183 and the p-value (two tailed) was <0.001. A subgroup analysis was performed based on gender, type of training and intensity. Aerobic training of moderate or high intensity has the highest potential to reduce visceral adipose tissue in overweight males and females. These results suggest that an aerobic exercise program, without hypocaloric diet, can show beneficial effects to reduce visceral adipose tissue with more than 30 cm2 (on CT analysis) in women and more than 40 cm2 in men, even after 12 weeks.

Overview of included studies for the meta-analysis: The initial search resulted in 107 articles after removing duplicates. After screening on title, abstract and full-text 15 articles (totalling 852 subjects in the exercise-only groups) fulfilled the a priori set inclusion criteria (Fig. 1). Because all studies that used dual energy x-ray absorptiometry or ultrasound did not fulfill one or more inclusion criteria, all of the 15 included studies used CT scan or MRI to quantify VAT.

Study quality: All of the 15 studies included in the qualitative analysis (fig 1) scored more than the a priori defined limit of 8/15 (range = 15 to 10), assessed with the “Critical Review Form–Quantitative studies” assessment instrument. In five of the fifteen studies, the reviewers independently agreed on all the items. For the other studies the degree of disagreement ranged between 1 and 5 items (mostly content validity related items). Disagreement was solved by consensus in all cases.

Meta-analysis: Data were extracted by a reviewer and presented in a spreadsheet to two other reviewers. The data were discussed until consensus was reached and analyzed in work meetings. All fifteen studies analyzed the effect of exercise on visceral adipose tissue (Table 1). The results of these studies were analyzed using the random effects model because of the high heterogeneity between studies (Fig 2). The standardized mean difference (Hedge's g) of the change in visceral adipose tissue after a physical activity intervention was −0.497 with a 95% confidence interval (95%CI) of −0.655 to −0.340. The Z-value was −6.183 and the p-value (two tailed) was <0.001. Heterogeneity analysis showed low to moderate, but significant, heterogeneity between studies (Cochran's Q = 37.317; degrees of freedom of Q (df(Q)) = 14; p = 0.001; I2 = 62.484).

Through a funnel plot of standard error by Hedge's g, a graphic representation of heterogeneity was obtained (Fig 3). Using the ‘fail’n safe' algorithm, it becomes clear that there was no publication bias in the analysis because 405 extra studies would be needed to get the p-value to the alpha level.

Three of the analyzed studies (Boudou 2001, Cooker 2009, Koo 2010) had small sample sizes (n<14) and could thereby overestimate the general effect of the therapy. Conducting a sensitivity analysis by removing these studies, still revealed a decrease in visceral adipose tissue of −0.464 Hedge's g after a physical activity intervention program (95%CI: −0.313 to −0.616, p<0.01). By making this advanced analysis, heterogeneity decreased, however only slightly, to Q = 28.741 (with p = 0.002, I2 = 61.727).

Subgroup analysis and Meta-Regression / Controlled versus uncontrolled: For the first subgroup analysis, the studies were divided in nine controlled and six uncontrolled trials (Fig 4). The analysis of these subgroups revealed that the controlled studies found a greater effect when comparing the results of VAT of the control group with those of the intervention group (Hedge's g = −0.561, 95%CI: −0.332 to −0.791, p<0.01) than uncontrolled studies in which the effect was compared between pre and post exercise intervention VAT values (Hedge's g = −0.437, 95%CI: −0.207 to −0.667, p<0.01).

Heterogeneity analysis showed significant moderate to high heterogeneity between controlled studies (Cochran's Q = 32.888; df(Q) = 8; p<0.001; I2 = 75.675) but homogeneity between uncontrolled studies (Cochran's Q = 2.469; df(Q) = 5; p = 0.781; I2 = 0.000).

Gender: In female studies the decrease in VAT was −0.550 Hedge's g (95%CI: −0.269 to −0.831, p<0.01) while in male studies the decrease in VAT was −0.589 Hedge's g (95%CI: −0.276 to −0.896, p<0.001). In two studies where there were mixed samples of males and females, a lower overall decrease of VAT was observed. (Hedge's g = −0.330, 95%CI: −0.045 to −0.706, p = 0.085). Heterogeneity analysis showed significant low to moderate heterogeneity in mixed (Cochran's Q = 10.021; df(Q) = 4; p = 0.040; I2 = 60.085), male (Cochran's Q = 22.007; df(Q) = 6; p = 0.001; I2 = 72.736) and female studies (Cochran’s Q = 25.878; df(Q) = 8; p = 0.001; I2 = 69.086).

By removing the data of the subjects who received strength training or mixed (aerobic and strength) training, heterogeneity between studies decreased, especially in female and mixed gender study populations (Cochran’s Q = 13.279; df(Q) = 6; p = 0.039; I2 = 54.815 vs. Cochran’s Q = 6.113; df(Q) = 2; p = 0.047; I2 = 67.282). The effect of physical activity stayed significant in both male (Hedge’s g = −0.545, 95%CI: −0.286 to −0.805, p<0.001) and female studies (Hedge’s g = −0.334, 95%CI: −0.083 to −0.585, p<0.001). However, in the mixed study groups, the effect of physical activity did not reach significance (Hedge’s g = −0.396, 95%CI: −0.054 to −0.847, p = 0.085).

Training Modality: All studies were categorized in “aerobic training”-studies, “strength training”-studies or “combined” subgroups. This subgroup analysis showed that the decrease in VAT in the aerobic training subgroup was higher (Hedge’s g = −0.550, 95%CI: −0.332 to −0.768, p<0.001) compared to the strength training subgroup (Hedge’s g = −0.529, 95%CI: −0.003 to −1.054, p = 0.049).

However, in the “combined” subgroups, the decrease in VAT was no longer significant (Hedge’s g = −0.301, 95%CI: 0.221 to −0.823, p = 0.258).

Heterogeneity between aerobic studies was significant low to moderate (Cochran’s Q = 39.121; df(Q) = 16; p = 0.001; I2 = 59.101) while significant low to high heterogeneity was observed between strength studies (Cochran’s Q = 23.824; df(Q) = 1; p<0.001; I2 = 95.803). Homogeneity was found between combination studies (Cochran's Q = 0.034; df(Q) = 1; p = 0.853; I2 = 0.000).

Training Intensity: A final subgroup analysis was made through the categorization for intensity of all aerobic studies in “high intensity”, “moderate intensity” and “low-intensity” studies (Fig 5). Following cut-off values were used: 1) “high intensity”: when training most of the time at an intensity≥70% of maximal heart rate (HRmax) or more than 55% of maximal oxygen consumption (VO2max) or 60–80% of the heart rate reserve (HRres); 2) “moderate intensity”: when training most of the time at an intensity between 60–70% HRmax or 45−55% VO2max and 3)“low intensity”: when training most of the time at an intensity<60% HRmax or <45% VO2max.

This analysis showed that a certain threshold of physical activity intensity should be reached to obtain a decrease in VAT. Only in studies in which moderate and high intensity-training was used, the effect on VAT was significant (“moderate intensity”-training: Hedge's g = −0.473, 95%CI: −0.140 to −0.806, p = 0.005; “high intensity”-training: Hedge's g = −0.588, 95%CI: −0.336 to −0.840, p<0.001).

Heterogeneity analysis showed significant low to moderate heterogeneity between “high intensity”-studies (Cochran's Q = 35.524; df(Q) = 10; p<0.001; I2 = 0.144) and homogeneity between “low intensity”-studies (Cochran's Q = 0.627; df(Q) = 1; p = 0.428; I2 = <0.001) and “moderate intensity”-studies (Cochran's Q = 2.659; df(Q) = 5; p = 0.751; I2 = <0.001).

The regression model showed no significant effect of duration of the intervention on VAT decrease ( = 0.836)

Re-expression of the estimate for clinical purpose: The five controlled studies (totalling 569 participants) using cm2 as the measuring scale of VAT showed an overall estimate of Hedge's g = −0.630 representing a pooled effects size of −37.1 cm2 (females), −46.5 cm2 (males) and −42.1 cm2 (combined gender groups) respectively.

Obesity is a worldwide public health problem. Currently, increasing evidence suggests acupuncture and related therapies are effective for obesity. This network meta-analysis (NMA) was performed to compare the effectiveness of different acupuncture and related therapies. We searched potential randomized controlled trials (RCTs) in three international databases. Thirty-four trials involving 2283 participants were included. Pairwise meta-analysis showed that acupuncture and related therapies were superior to lifestyle modification and placebo in reducing weight and body mass index (BMI). Based on decreases in body weight, results from NMA showed that acupoint catgut embedding (standard mean difference [SMD]: 1.26; 95% credible interval [95% CI], 0.46–2.06), acupuncture (SMD: 2.72; 95% CrI, 0.06–5.29), and combination of acupuncture and related theories (SMD: 3.65; 95% CrI, 0.96–6.94) were more effective than placebo. Another NMA result indicated that acupoint catgut embedding (SMD: 0.63; 95% CI, 0.25–1.11), acupuncture (SMD: 1.28; 95% CrI, 0.43–2.06), combination of acupuncture and related therapies (SMD: 1.44; 95% CrI, 0.64–2.38), and electroacupuncture (SMD: 0.60; 95% CrI, 0.03–1.22) were superior to lifestyle modification in decreasing BMI. Combination of acupuncture and related therapies was ranked the optimal method for both reducing weight and BMI. Further studies will clarify which combination of acupuncture and related therapies is better.

3.1. Study Search: We performed this research on Dec 26 2017. As shown in Figure 1, a total of 1050 records were initially identified from the databases. 675 studies left after duplicates were removed. 577 records were excluded after carefully scanning titles and abstracts. Finally, 34 trials with 2283 participants were included in our NMA, covering 8 groups, manual acupuncture; electroacupuncture; auricular acupoint stimulation; acupoint catgut embedding; pharmacotherapy; warming acupuncture; lifestyle modification; placebo.

3.2. Study Description: Main characteristics of included RCTs were shown in Table 1. The participants were from Australia, the United States, Turkey, Korea, Iran, Egypt, and China. Age of participants ranged from 15 to 70 years, while the sample size of the studies ranged from 12 to 86. Among the included RCTs, there were one four-arm trials, 5 three-arm trials, and 28 two-arm trials. Fourteen studies compared acupuncture to placebo. Ten studies compared acupuncture to lifestyle intervention. Six studies compared combined therapies to acupuncture alone. Details about acupuncture points used, retention time, frequency, and duration of acupuncture sessions were shown in Table 2. In these research, 30 articles reported the weight loss, while 25 articles reported the change in BMI. The details of mean, standard difference (SD), and sample size between different groups for eligible studies are summarized in Appendix 3. The Cochrane risk of bias assessment was presented in Table 3. Furthermore, the network plot of included comparisons was shown in Figure 2.

3.3. Pairwise Meta-Analyses / 3.3.1. Body Weight: A direct pairwise meta-analysis showed that acupuncture and related therapies showed a greater BW reduction than lifestyle modification (MD: 1.66; 95% Confidence interval, 0.63to2.70) and placebo (MD: 1.15; 95% CI, 0.67to1.63). When compared to acupuncture, combination of acupuncture and related theories showed a marginally stronger effect in weight loss (MD: 1.56; 95% CI, 0.07to3.05). There was no statistically significant difference between combination of acupuncture and related theories and pharmacotherapy in their effectiveness in BW (MD: 2.44; 95% CI, -1.98to6.86). (Table 4)

BMI. As for the comparison in reducing BMI, acupuncture and related therapies were found to be marginally superior to lifestyle modification (MD: 1.17; 95% CI, 0.09to2.26) and placebo (MD: 0.57; 95% CI, 0.40to0.74). The remaining direct comparisons did not show significant differences (Table 4).

3.4. Network Meta-Analysis / 3.4.1. Body Weight: The NMA showed that all treatments other than acupuncture combined lifestyle modification were more efficacious than lifestyle modification. Three treatments were significantly more effective than placebo. Specifically, acupoint catgut embedding (SMD: 1.26; 95% credible interval, 0.46to2.06), acupuncture (SMD: 2.72; 95% CrI, 0.06to5.29), and combination of acupuncture and related therapies (SMD: 3.65; 95% CrI, 0.96to6.94). Furthermore, moxibustion with warming needle was associated with a significantly improvement than lifestyle modification (SMD: -5.24; 95% CrI, -10.15to-0.55) (Table 5).

3.4.2. BMI: Four treatments showed superiority over placebo, including acupoint catgut embedding (SMD: 1.31; 95% CrI, 0.36to2.06), acupuncture (SMD: 1.94; 95% CrI, 0.83to3.00), combination of acupuncture and related theories (SMD: 3.65; 95% CrI, 0.96to6.94), and electroacupuncture (SMD: 1.28; 95% CrI, 0.43to2.11). Four treatments were significantly more effective than lifestyle modification, including acupoint catgut embedding (SMD: 0.63; 95% CI, 0.25to1.11), acupuncture (SMD: 1.28; 95% CrI, 0.43to2.06), combination of acupuncture and related theories (SMD: 1.44; 95% CrI, 0.64to2.38), and electroacupuncture (SMD: 0.60; 95% CrI, 0.03to1.22). Also, the combination of acupuncture and related theories and acupuncture alone were both superior to the acupuncture combined lifestyle modification in their ability to reduce body mass index (SMD = -1.76, 95% CrI =−2.96 to −0.62; SMD = −1.59, 95% CrI = −2.71to −0.34) (Table 5).

3.5. Ranking / 3.5.1. Body Weight: Ranking of the different treatment methods was displayed Figure 3. The results suggested that, on the aspect of weight loss, combination of acupuncture and related therapies was ranked the optimal method, the best, (88.7%), followed by moxibustion with warming needle (87.8%), manual acupuncture (70.5%), acupoint catgut embedding (ACE,62.1%), auricular acupoint stimulation (AAS,48.3%), electro acupuncture (EA,46.3%), pharmacotherapy (41.9%), acupuncture combined lifestyle modification (AR+LM,31.2%), placebo acupuncture/sham acupuncture (16.8%), and lifestyle modification (LM,6.4%) which was ranked as the worst.

3.5.2. BMI: The results suggested that, on the aspect of BMI, combination of acupuncture and related therapies was ranked the optimal method, the best, (90.2%), followed by manual acupuncture (83.3%), pharmacotherapy (64.7%), acupoint catgut embedding (58.6%), auricular acupoint stimulation (55.7%), electroacupuncture (52.1%), placebo acupuncture/sham acupuncture (25.1%), acupuncture combined lifestyle modification (16.9%), and lifestyle modification (3.4%) which was ranked as the worst.

3.6. Inconsistency Assessment / 3.6.1. Body Weight: The Z test illustrates the inconsistency of the NMA specifically (Appendix 4). For the inconsistency test outcome of BW, 95% CI of 8 loops was included 0, which reflected that no significant inconsistency was found. However, another 2 loops (ACE-Acupuncture-Combined therapies; ACE-Combined theories-EA) were found statistical inconsistency between direct and indirect comparisons.

3.6.2. BMI: For the inconsistency test outcome of body mass index, 95% CI of all loops (acupuncture -combined theories-EA; acupuncture-EA-placebo; AR+LM -EA- LM-placebo; ACE-acupuncture-EA; AAS-EA-LM; AAS-AR+LM-LM-placebo; AAS-EA-placebo) were included 0, which reflected that no significant inconsistency was found.

3.7. Safety: Ten RCTs reported adverse events, while no major complications were noticed in all included studies. Three included studies reported that no adverse effects were noted in both experimental group and placebo group. In one included RCT, there were two patients reporting mild ecchymosis and one abdominal discomfort case reported as adverse events after electroacupuncture treatment; no case was reported in the lifestyle modification group. In another study, there were seven subjects in group auricular acupoint stimulation and two subjects in group placebo had mild tenderness.

Herbal medicine based on pattern identification (PI) is used widely in Traditional Chinese Medicine. Proper herbal medicine based on PI has been suggested for effective weight reduction and decreasing the adverse events. This systematic review examined the effectiveness and safety of herbal medicine, focusing on PI in treating obesity.

Methods: Eight electric databases were used for searching randomized controlled trials (RCT) (to August 31, 2021). RCTs which prescribed herbal medicine to obese patients based on PI were included. Body weight (BW) and body mass index (BMI) were the primary outcomes. The risk of bias was assessed using Cochrane risk of bias tool, and the meta-analysis was conducted. Grading the evidence was conducted by using GRADEpro.

Results: Sixteen RCTs (1052 patients) were included: 2 studies compared herbal medicine to placebo (128 patients); 2 studies compared them to western medication (161 patients); 12 studies compared them with usual care (763 patients). The meta-analysis showed that the herbal formulas reduced the BW and BMI without significant Adverse events compared to the control group (BW: mean difference = –4.10, 95% confidence interval: –5.14 to –3.06, I2 = 2% and BMI: mean difference = –1.53, 95% confidence interval: –1.88 to –1.19, I2 = 25%). Moderate-quality evidence on the primary outcomes was found.

Conclusions: Herbal medicine – has good clinical efficacy and safety in treating obesity. This study has limitations that some literatures with high risk of bias in blinding or without using a standardized diagnosis of PI were included. However, the current evidence suggests the possibility of precision medicine using PI.

3.1. Study inclusion: A total of 2932 citations were identified. Studies that did not use PI or inappropriate subjects, comparators, or data were excluded. Finally, 16 RCTs were included (1052 patients). According to the comparison, 2 RCTs (128 patients) compared herbal medicine to a placebo; 2 RCTs (161 patients) compared them to Western medication; 12 RCTs (763 patients) compared them to usual care, including modulation of the diet or exercise (Fig.1).

3.2. Study characteristics: Table 1 lists the characteristics of the included studies (16 RCTs). All studies included were performed in China. The number of patients involved in the studies ranged from 25 to 40 in the treatment or control groups. In total, 14 kinds of herbal formulas were involved. The forms of the intervention included decoction (12 RCTs), capsule or tablet (4 RCTs), and powder (1 RCT). Two RCTs used a placebo as a comparator, and 2 RCTs used orlistat for comparison. Twelve RCTs used diet and exercise control as a comparator of usual care. Most included reasonable diets or limited calories and aerobic exercise. The treatment duration ranged from 28 days to 3 months; 12 weeks (3 months) was mostly used (8 RCTs), followed by 8 weeks (2 months or 60 days) (6 RCTs). Modulation of diet and exercise was used in 15 RCTs. Only 2 studies used both BMI and BW as the outcome measure. Nine, 6, and 7 RCTs used WC, HC, and WHR, respectively, as an outcome measure. Eight RCTs reported the result of AE, and most were not significant in either group.

3.3. Pattern identification: Table 2 lists the classification of PI and the diagnostic criteria used for PI.

The studies were assorted into 3 major types according to the pathology; the phlegm-dampness type, the heat accumulation type, and the liver-qi stagnation type. The phlegm-dampness type was counted in 8 RCTs, and it included only phlegm-dampness (n = 2) and phlegm-dampness with spleen deficiency (n = 6). The heat accumulation type was counted in 6 RCTs and commonly, it was related to the spleen and stomach. The type included stagnation of heat (or damp-heat) in the spleen and stomach, stomach heat and dampness stagnation, and damp-heat accumulation. The liver-qi stagnation type was collected in 2 RCTs and called a spleen deficiency and liver-qi stagnation or spleen deficiency and stagnation of liver-qi and heat.

‘Diagnosis and efficacy evaluation criteria of simple obesity’ was most frequently used in Seven RCTs as the diagnostic criteria for PI. ‘Internal Chinese medicine’, ‘Guiding principles for clinical research on the new drug of traditional Chinese medicine’, and ‘Guidelines for diagnosis and treatment of common diseases in internal Chinese medicine’ were all used in 3 RCTs, respectively. ‘Endocrinology specialty diseases and rheumatism-Clinical diagnosis and treatment of Chinese medicine, 2nd Edi.’ was used in 2 RCTs. Three RCTs used 2 kinds of diagnostic criteria. “Guidelines for diagnosis and treatment of common diseases in internal Chinese medicine” and “Internal Chinese medicine” were used in 2 RCTs, and “Guiding principles for clinical research on the new drug of traditional Chinese medicine” and “Diagnosis and efficacy evaluation criteria of simple obesity” were used in 1 RCT.

3.4. Risk of bias: Fig. 2 indicates the risk of bias. In relation to selection bias, random sequence generation was low in 12 RCTs and unclear in 4. Allocation concealment was low in 3 RCTs and unclear in 13. Blinding of the participants and personnel was a high risk of bias in most studies, except for 2 RCTs that compared herbal medicine with a placebo. Another domain related to blinding and detection bias presented a low risk only in 1 RCT. Other RCTs did not describe sufficient information to assess the risk of bias. The risk of bias in incomplete outcome data was low in 11 RCTs and unclear in the others. The risk of selective reporting was assessed to be low in 5 RCTs; the risk of bias in the others was unclear, as there was a lack of explanation to assess the risk of bias. Most studies did not describe the information needed to assess the risk of bias; the other bias was also assessed as unclear. Only 2 RCTs whose comparator was placebo were assessed as having a low risk of bias overall. The others had a low or unclear risk of bias.

3.5. Outcomes / 3.5.1. BW and BMI.: Herbal medicine based on PI led to a significant reduction in both BW and BMI (BW: mean difference [MD] = –4.10, 95% confidence interval [CI]: –5.14 to –3.06, P <.0001, I2 = 2%, BMI: MD = –1.53, 95% CI: –1.88 to –1.19, P <.0001, I2 = 25%).

When herbal medicine was compared with comparators, the changes in BMI were statistically significant in all subgroup analyses. In subgroup analysis, however, which compared herbal medicine to placebo, the decrease in BW was not significant in the herbal medicine group compared to the placebo group (BW: MD = –4.00, 95% CI: –10.52 to 2.52, P =.23, I2 = not applicable). (Fig.3)

3.5.2. WC, HC, and WHR.: The meta-analysis found that herbal medicine based on PI induced a significant decrease in WC, HC and WHR (WC: MD = –2.48, 95% CI: –2.95 to –2.02, P <.00001, I2 = 0%, HC: MD = –1.75, 95% CI: –3.21 to –0.29, P =.02; I2 = 65%, and WHR: MD = –0.03, 95% CI: –0.05 to –0.01, P =.0003; I2 = 80%).

After subgroup analysis according to comparator including placebo, orlistat, and usual care, herbal medicine induced a significant decrease only when compared with usual care. Herbal medicine could not lead to a significant improvement in WC and HC compared to the placebo. (Fig. 4)

3.6. Adverse events: Eight RCTs did not report AEs, whereas the other 8 RCTs reported the safety of herbal medicine. Only 2 studies reported AEs in the experimental group. On the other hand, they were mild AEs (4 cases), such as diarrhea, and they did not cause severe results.

3.7. Publication bias: Fig.5 presents the publication bias according to the funnel plot. The funnel plot was considered visually asymmetric; hence, it was inferred that the publication bias possibly exists.

3.8. Assessment of evidence: The quality of evidence for the primary outcome, BW and BMI, was assessed as moderate with a high risk of blinding. The assessment of the evidence for WC was also presented moderate quality. On the other hand, the quality of evidence for HC and WHR was downgraded as “very low” and “low,” respectively, because of the serious inconsistency (high heterogeneity) and imprecision (small sample size).