Effect of Low-Carbohydrate vs Low-Fat Diet Intervention on Visceral Fat in a 12-Month Randomized Controlled Trial

doi:10.21203/rs.3.rs-4926524/v1

Download PDF

Article

Effect of Low-Carbohydrate vs Low-Fat Diet Intervention on Visceral Fat in a 12-Month Randomized Controlled Trial

https://doi.org/10.21203/rs.3.rs-4926524/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Dietary restriction via a healthy low-fat (HLF) diet or a healthy low-carbohydrate (HLC) diet vary in their effects on adiposity and metabolism. The HLC diet, but not HLF diet, may preferentially reduce visceral adipose tissue (VAT), the major adipose tissue contributing to metabolic deregulation.

In a 12-month weight loss trial, DIETFITS (Diet Intervention Examining The Factors Interacting with Treatment Success), we compared VAT loss between HLF and HLC diets by randomizing adults to either diet. VAT was measured using dual-energy x-ray absorptiometry. Linear mixed models analyzed associations between diet and VAT.

Among 449 participants (60% women; mean age 39 years), VAT loss was significantly greater for those eating the HLC diet compared to the HLF diet at 6 months [10.6cm²; 95% confidence interval (CI): 5,16.2] and 12 months (6.3cm²; 95% CI: 0.6,12). Preferential VAT loss was greater in participants eating the HLC diet at 6 months only. Men experienced greater HLC diet-induced VAT loss than women. Insulin secretion status did not modify VAT loss.

HLC diet reduced metabolically harmful VAT, particularly during the first 6 months of diet. Lowering VAT has the potential to reduce risk for cardiometabolic disease. Sex differences should be considered in designing effective dietary interventions.

Health sciences/Risk factors

Health sciences/Medical research/Clinical trial design/Randomized controlled trials

Macronutrient restriction via a low-fat (LF) diet and a low-carbohydrate (LC) diet are common strategies for weight loss that may vary in their effects on fat and metabolism. Past trials comparing weight loss effectiveness with LF and LC diets have been inconsistent.¹ In the Diet Intervention Examining The Factors Interacting with Treatment Success (DIETFITS) trial, we previously found no weight loss differences between a healthy LF (HLF) and a healthy LC diet (HLC) after 12-months.² Other LC trials reported rapid initial weight loss³, but this may have been driven by loss of water weight and lean body mass rather than fat loss.¹ Although total weight loss is easily measured as a trial endpoint, visceral adipose tissue (VAT) loss is more difficult to measure but more informative. Beyond subcutaneous adipose tissue (SAT), VAT plays a central role by which obesity deregulates energy metabolism—the rate at which the body burns energy—within the cardiometabolic disease sequelae..⁴ Thus, a diet study measuring decreased VAT levels would be ideal to evaluate overall cardiometabolic improvement.

How VAT is mobilized could be key to differentiate between LC and LF diets. The LC diet appears to preferentially decrease VAT by lowering postprandial glucose.⁵ Carbohydrates elevate blood glucose levels, stimulating insulin secretion, which promotes lipogenesis and preferential visceral deposition.⁶ A systematic review of weight loss interventions reported that VAT was preferentially lost during early and/or modest weight loss but attenuated in the long-term.⁷ Accordingly, given this biological understanding and accumulating clinical evidence, directly measuring VAT as the trial endpoint would more accurately compare differences in weight loss between LC and LF diets and help explain the mixed results from past weight loss trials.⁴

Physiological differences in diet response through insulin resistance and sex may moderate LC and LF diet efficacy. There is mixed evidence supporting this, with some studies suggesting that maintaining a LC diet may be superior to LF for individuals with higher insulin secretion or higher insulin resistance.^1,8 Conversely, DIETFITS found no association by diet and baseline insulin secretion status and weight loss after 12 months.² While most trials use a weight loss endpoint, the role of insulin resistance likely modulates VAT directly by lowering demand on insulin-mediated glucose disposal. A non-randomized trial found that insulin-resistant participants in the LC group experienced greater VAT loss than the LF group over 15 weeks.⁹ There is a need to test this hypothesis in a randomized trial with longer-term follow-up. In addition, sex also modifies fat deposition and mobilization, with men preferentially depositing VAT over SAT. A secondary analysis of DIETFITS found that men experience greater weight loss eating a HLC diet compared with women.¹⁰ VAT may be a mechanism underlying sex and insulin secretion dynamics in LC diet efficacy.

To understand whether the HLC and HLF diets differ in VAT loss, we conducted a randomized analysis with derived primary data on VAT and SAT at three timepoints in the DIETFITS trial. Our overarching objective was to measure the differences in VAT loss in participants eating a HLF diet or a HLC diet in a long-duration study (Aim 1) and determine any interactions between diet and sex or insulin secretion status (Aim 2). For Aim 1, we hypothesized that a HLC diet would be associated with greater VAT loss compared with a HLF diet after 6 and 12 months. For Aim 2, we hypothesized that men and those with greater baseline insulin secretion assigned to HLC diet would be predisposed to a greater VAT loss.

Participant characteristics

There were 531 DIETFITS participants from cohorts 2–5 who were eligible for dual-energy x-ray absorptiometry (DXA) scans, and 449 participants in the final analysis set after excluding those who did not receive scans at baseline. Twelve VAT areas were below the lower limit of detection and set to be 50; similarly, one SAT area was below the limit and set to 50. Baseline characteristics are presented in Table 1 by arm. We observed minimal to small differences between HLC and HLF groups based on absolute standardized differences (ASDs: 021-0.209). Mean VAT, SAT, and VAT/SAT ratio are displayed by diet arm and timepoint (Fig. 1).

Table 1

Baseline participant characteristics compared using the absolute standard differences (ASD) between the healthy low-carbohydrate (HLC) and healthy low-fat (HLF) diet arms
	HLC	HLF	ASD
	N = 228	N = 221
Sex, N (%)			0.080
Men	86 (37.7)	92 (41.6)
Women	142 (62.3)	129 (58.4)
Age, mean years (SD)	39.41 (6.78)	38.91 (6.71)	0.073
Race/Ethnicity, N (%)			0.085
Hispanic	48 (21.2)	52 (23.6)
NH Asian	22 ( 9.7)	20 ( 9.1)
NH Black	10 ( 4.4)	7 ( 3.2)
NH Other	12 ( 5.3)	12 ( 5.5)
NH White	134 (59.3)	129 (58.6)
Marital Status, N (%)			0.140
Married/partnered	135 (59.5)	145 (65.9)
Single/never married	69 (30.4)	54 (24.5)
Divorced/separated/widowed	23 (10.1)	21 ( 9.5)
Highest education level, N (%)			0.209
College	129 (56.8)	123 (55.7)
High School	9 ( 4.0)	2 ( 0.9)
Post-graduate	89 (39.2)	96 (43.4)
Baseline insulin 30, mean (SD)	89.33 (61.33)	90.69 (66.88)	0.021
VAT, mean (SD)	166.35 (59.21)	162.16 (54.34)	0.074
SAT, mean (SD)	451.70 (100.80)	460.98 (118.97)	0.084
Lean mass, mean (SD)	12844.87 (2671.81)	13046.76 (2343.72)	0.080

Aim 1. VAT loss in HLC vs. HLF participants at 6 and 12 months

Based on our primary model, we observed a significant difference in VAT loss between HLF and HLC diets at both 6 and 12 months (Fig. 2). On average, compared to participants eating the HLF diet, participants eating the HLC diet lost 10.60 cm² [95% confidence interval (CI): 5.00, 16.21] more VAT at 6 months from baseline. At 12 months from baseline, compared to participants eating the HLF diet, participants eating the HLC diet lost 6.30 cm² (95% CI: 0.57, 12.03) more VAT.

After controlling for SAT in the secondary model, we observed a significant difference between diets only at 6 months from baseline, with participants eating the HLC diet losing 7.2 cm² (95% CI: 2.7, 11.87) more VAT compared to participants eating the HLF diet (Table 2). However, at 12 months from baseline participants eating the HLC diet lost 3.9 cm² (95% CI: -0.7, 8.6) more VAT than participants eating the HLF diet. These findings suggested that adjusting for SAT loss in the model slightly attenuates the difference in VAT loss between diets sometime between 6 and 12 months. While mixed-effects models are relatively robust to violations in distributional assumptions, this secondary model exhibited a mild violation of homoscedasticity, based on quantile deviations in the plot of residuals against predicted values.¹² Conclusions did not change in our sensitivity analysis using log transformation of VAT (Appendix 1), therefore, we present non-transformed results.

We also evaluated possible differences between diets for change in the VAT/SAT ratio over time. For this, we applied a log transformation of the ratio that improved model assumptions. Based on this model, we did not observe any significant difference in the change of VAT/SAT ratio between diets at 6 and 12 months. On average, compared to participants eating the HLC diet, participants eating the HLF diet experienced a 2.5% increase (95% CI: 0.996,1.055) at 6 months and a 0.7% increase (95% CI: 0.98,1.04) at 12 months in the VAT/SAT ratio. Boxplots in Fig. 1 display how both VAT and SAT decreased over time together and thus the ratio is approximately constant over time.

Table 2

Change estimates in mean visceral adipose tissue (VAT) area (cm²) by healthy low-fat (HLF) diet or healthy low-carbohydrate (HLC) diet
	HLF¹ N = 221	HLC¹ N = 228	HLF - HLC²	95% CI
Primary model VAT (cm²)
6 months	-19.42	-30.03	10.60	5.00, 16.21
12 months	-12.68	-18.98	6.30	0.57, 12.03
Men³
6 months	-26.59	-42.41	15.82	5.39, 26.25
12 months	-17.02	-29.18	12.16	1.60, 22.71
Women³
6 months	-14.55	-22.73	8.18	2.16, 14.20
12 months	-9.79	-12.95	3.16	-3.03, 9.35
Secondary model VAT controlling for SAT⁴
6 months	-8.58	-15.8	7.21	2.66, 11.77
12 months	-3.65	-7.56	3.91	-0.74, 8.57
Men³
6 months	-8.06	-19.79	11.73	3.54, 19.93
12 months	-4.69	-11.82	7.13	-1.18, 15.43
Women³
6 months	-4.86	-9.20	4.34	-0.45, 9.13
12 months	-0.32	-1.76	1.44	-3.47, 6.35
¹ Change estimates in mean VAT area from baseline to 6 and 12 months
² Between-group difference in VAT loss and 95% confidence interval (CI)
³ Sex-stratified models
⁴ Secondary model for mean VAT area additionally controlling for subcutaneous adipose tissue (SAT)

Aim 2. Effect modification by insulin secretion status and sex

To be consistent with the parent DIETFITS study and evaluate the effect modification of insulin secretion on the diet differences in VAT loss over time, we tested the 3-way interaction in a saturated version of the primary model. This interaction was not statistically significant (P = 0.39). Based prior evidence for effect modification by sex on the diet difference in VAT loss over time in DIETFITS, we presented stratified versions of the primary model by male and female (Table 2). Among men, VAT loss from baseline was significantly greater among those eating the HLC compared to the HLF diet at both 6 months (15.82cm², 95% CI: 5.39, 26.25) and 12 months (12.16cm²; 95% CI: 1.60, 22.71). Similarly, female participants eating the HLC diet observed significantly, 8.18 cm² (95% CI: 2.16, 14.20), greater VAT loss at 6 months compared to those eating the HLF diet. However, this difference attenuated to 3.16 cm² (95% CI: -3.03, 9.35) by 12 months. The greater VAT loss in the HLC compared to HLF diets was nearly 4 times greater among men (12.16 cm²) than among women (3.16 cm²). Although our secondary models adjusting for SAT loss displayed similar conclusions, among women we no longer observed a significant 6-month diet difference.

In this randomized clinical trial, our results support our primary hypothesis that compared to the HLF diet, the HLC diet was more effective in decreasing VAT at both 6 and 12 months from baseline. This suggests that the HLC diet is associated with favorable changes in adipose tissue through depletion of the metabolically harmful visceral depot. A diet that preferentially targets VAT would be an effective method of secondary and tertiary cardiometabolic disease prevention without the need for overall weight loss because VAT deregulates energy metabolism to a greater extent than other adipose types.⁴ Controlling for SAT demonstrated that the preferential loss VAT over SAT wanes after 6 months of dieting. Our findings also support our secondary aim showing that the observed advantage in the HLC diet was greater among men than women. However, baseline insulin secretion status did not modify diet efficacy.

Consistent with our primary hypothesis, the HLC diet resulted in a 1.5-fold greater reduction in VAT compared to the HLF diet over a 12-month period. Most previous HLC diet trials identified greater weight loss than the control diet in the short-term but have not distinguished whether the weight loss was driven by loss of fat, water weight, or lean body mass.¹ Our results extend on prior VAT loss trials that were shorter duration or not RCT designs. One 8-week RCT found 3-fold greater VAT loss in the LC diet compared to the LF diet arm,⁵ consistent with results from another 15-week non-randomized trial.⁹ However, other trials have found no diet difference in VAT loss.¹³ In agreement with clinical data, biological evidence suggests that lower postprandial glucose and insulin secretion following LC meals may preferentially decrease VAT. Compared with SAT, metabolic activity in VAT has nearly double the insulin-stimulated glucose uptake rate, which is amplified by reduced insulin secretion with a HLC diet.⁶

The HLC and HLF diets may differ in trajectories of lipolysis. As most VAT loss diet trials have not considered relative SAT loss, there is no consensus on method. While some studies have calculated the VAT/SAT ratio to index SAT, the use of index ratios in obesity research poses several analytical and interpretative issues.¹⁴ We found that controlling for SAT produced a more stable and interpretable estimate compared with the log transformed VAT/SAT ratio.

We found that preferential VAT loss (controlling for SAT) was 2-fold greater in the HLC diet than the HLF diet at 6 months, with the difference slightly attenuating by 12 months. Previous shorter-term HLC trials observed rapid initial weight loss that attenuates with time, but long-term trials are sparse. In DIETFITS, we previously found no significant diet differences in overall weight loss after 12 months. In the current study, distinguishing VAT from SAT provided a nuanced understanding of this weight loss trajectory. VAT has greater potential for initial lipolysis in response to HLC diet because visceral depot fatty acids have a higher turnover rate and are generally mobilized first during negative energy balance.⁴ Similar to our results, one systematic review found that greater VAT relative SAT loss occurred early on but was not sustained long-term (12–14 weeks).⁷ Taken with prior evidence and the parent DIETFITS study, our results suggest that the rapid initial weight loss at 6 months was promoted by the preferential loss of adipose tissue in the visceral depot relative to the subcutaneous depot, but after 6 months, this preferential VAT loss diminished during sustained weight loss.

This study provides evidence that the trajectory of VAT loss from the HLC diet differed by sex, with men in the HLC arm losing 6% more VAT than women and sustaining significant VAT loss over 12 months. Furthermore, the HLC diet preferentially decreased VAT relative to SAT among men but not women. This aligns with a prior DIETFITs report that found a gender-based sociocultural preference for LC diets with men having a greater preference and significantly more adherent to a LC diet, but women expressing more preference for LF foods.¹⁰ This diet-gender preference is amplified biologically by preferential adipose accumulation and lipolysis in the visceral depot in men.¹⁵ In contrast, estrogens modulate adipose accumulation to the subcutaneous depot among women.¹⁵ Despite this biological plausibility, few diet trials have investigated sex differences in VAT loss. One 3-month LC diet trial connected a reduction in dietary carbohydrate intake with significant correlations to VAT loss among men but not women.¹⁶ Given the importance of VAT loss strategies to cardiometabolic disease reduction, we need to understand gendered approaches to improving adherence to LC diets towards investigating the physiological role of LC diets on VAT among women.

We hypothesized that the HLC diet would improve VAT loss among individuals with higher insulin resistance, but we found no evidence for diet-insulin interaction. The LC diet is thought to lower demand on insulin-mediated glucose disposal for those with impaired insulin metabolism, permitting greater oxidation of fatty acids.⁸ Fatty acids from the visceral depot have a higher turnover rate, and are generally mobilized first during negative energy balance, which may be amplified by the reduction in insulin secretion with a LC diet.⁵ Prior findings evaluating the role of insulin resistance in diet response have produced mixed results. DIETFITS provided strong evidence of null effects on weight loss in a large sample size and long-duration trial. Our findings extend these null findings by directly analyzing adipose sub-types.

Our results should be interpreted with consideration of limitations, which were also reported in the parent study.² Study participants had higher educational attainment and income, which may have improved retention and diet adherence compared with the general population. DIETFITS participants represented similar composition to the US across White, Hispanic, Asian, which may not be generalizable to African American or American Indian or Alaska Native groups. Given that the identified differences may be heavily influenced by estrogen, it may not be generalizable to postmenopausal women. More research is needed to further disentangle whether differences are related to biological sex differences (e.g., estrogens), sociocultural gender differences (e.g., low-fat diet preference in women), or a combination of both. While our analysis considered the role of physiological factors, regulation of body composition also involves a complex interaction with behavioral and environmental factors.

This study had several strengths contributing to the robustness of its evidence. Directly and longitudinally measuring VAT and SAT using a method comparative to the gold-standard CT offered unique insights into the adipose mechanics during diet-induced weight loss. The long duration and large sample size of DIETFITS also offered robust evidence that built on previous trials with small sample sizes and shorter duration trials. Participant weight loss during our study was significant and allowed for the opportunity to meaningfully test the diets. Adherence was good on both diets, and diet quality was more correlated with participant weight loss than calorie restriction during the first 6 months.¹⁷

In conclusion, VAT loss is a primary mechanism in preventing and reversing cardiometabolic disease and accordingly, targeting VAT loss has important clinical implications for primary and secondary prevention of cardiometabolic disease. In our analysis, the HLC diet was associated with greater VAT loss compared to the HLF diet. Our results extend on the findings of several smaller studies that identified a trajectory of rapid initial VAT loss in the short-term (weeks), by providing evidence of this trajectory through 1 year.⁷ The short-term effectiveness of the HLC diet in VAT loss also has clinical application for patients at high cardiometabolic risk in need of rapid VAT reduction. We identified sex-specific diet efficacy with men achieving greater health benefits from the HLC diet at 6 and 12 months, while the diet-difference in VAT loss diminished among women by 12 months. Given the consistency of evidence across the literature for the short-term nature of HLC diet benefits, future HLC study designs evaluating long-term and sex differences are needed.

DIETFITS was a randomized trial of adults aged 18–50 years with a BMI between 28 and 40 and without diabetes in a parallel group design to evaluate the difference in weight loss at 12 months for participants eating HLC or HLF diets. The primary outcome for the parent study was 12-month weight change. The DIETFITS protocol and primary outcomes have been described previously.² The Stanford Institutional Review Board approved all study procedures. All study participants provided written informed consent.

Study population

The parent DIETFITS study recruited 609 adults from the San Francisco Bay Area of California using media advertisements and email lists between March 2013 and March 2015. Eligibility criteria included men and premenopausal women aged 18 to 50 years with BMI of 28 to 40. The main exclusion criteria were having uncontrolled hypertension or metabolic disease, diabetes, cancer, or heart, renal, or liver disease or being or becoming pregnant or lactating. Individuals were also excluded if taking hypoglycemic, lipid-lowering, antihypertensive, psychiatric, or other medications known to affect body weight or energy expenditure. Participants were randomized to a HLF diet or a HLC diet using an allocation sequence determined by computerized random-number generation (Blockrand in R version 3.4.0; R Project for Statistical Computing) in block sizes of 8 (with 4 individuals going to each diet) by a statistician not involved in intervention delivery or data collection. Five cohorts had staggered start dates between March 2013 and March 2015. Additional funding near the onset of the study allowed for the assessment of DXA in cohorts 2 through 5 (i.e., participants in cohort 1 were not assessed). The current analysis focuses on these participants with DXA measures.

The protocol included a 1-month run-in period during which participants were instructed to maintain their habitual diet, physical activity level, and body weight. The intervention involved 22 instructional sessions held over 12 months in diet-specific groups of approximately 17 participants per class. Sessions were held weekly for 8 weeks, then every 2 weeks for 2 months, then every 3 weeks until the sixth month, and monthly thereafter. Classes were led by 5 registered dietitian health educators who each taught a HLF class and HLC class per cohort. Dietitians were blinded to all laboratory measures and genotype. The dietary interventions were described previously.² Briefly, the main goals were to achieve maximal differentiation in intake of fats and carbohydrates between HLC and HLF diets while otherwise maintaining equal treatment intensity and an emphasis on high-quality foods and beverages. Thus, participants were instructed to reduce intake of total fat or digestible carbohydrates to 20 g/d during the first 8 weeks. Higher priorities for reduction were given to specific foods and food groups that derived their energy content primarily from fats or carbohydrates. Then individuals slowly added fats or carbohydrates back to their diets in increments of 5–15 g/d per week until they reached the lowest level of intake they believed could be maintained indefinitely. No explicit instructions for energy (kilocalories) restriction were given. Both diet groups were instructed to (1) maximize vegetable intake; (2) minimize intake of added sugars, refined flours, and trans fats; and (3) focus on whole foods that were minimally processed, nutrient dense, and prepared at home whenever possible.² All participants were also instructed to follow their current physical activity recommendations. Before randomization and at months 6 and 12, an oral glucose tolerance test of 75 g was administered, including measurement of insulin concentration 30 minutes after glucose consumption (i.e., INS-30, a proxy measure of insulin secretion).

Outcome assessment

Staff who measured outcomes were blinded to diet assignment. Whole body DXA scans were conducted among cohorts 2–5 to measure total body fat and muscle at 0, 6, and 12 months (Hologic model QDR densitometer; Hologic, Inc., Waltham, MA). In 2021, DXA scans were reanalyzed to generate VAT and SAT area estimates using the APEX 4.0 software (Hologic Inc., Bedford, MA). We previously validated our approach, finding that APEX provided highly valid VAT and SAT measures compared to MRI, the gold standard, resulting in a correlation coefficient of 0.90 and 0.92, respectively.¹¹ APEX is a semi-automated software using algorithms to estimate VAT and SAT in the abdominal region of interest, 5 cm in height across the full width of the abdomen at approximately the 4th lumbar vertebrae. APEX assigned lines of demarcation in the abdominal region of interest on both the outer edges of the soft tissue and the visceral cavity area. Algorithms generated values for VAT and SAT area (cm²): subtracting the derived SAT value from total abdominal adipose tissue estimate results in an estimate of VAT. We assumed 50 cm² was the lower limit of detection, and values below this were replaced with the lower limit.

Statistical analysis

Baseline characteristics were summarized by mean (standard deviation) or n (percent) for continuous and categorical variables, respectively. We compared characteristics by arm for HLF (N = 221) and HLC (N = 228) using ASD, for which a value of 0.2, 0.5, or 0.7 corresponds to small, medium, or large differences at baseline between arms.

To evaluate Aim 1, the change in VAT after 6 and 12 months eating HLF and HLC diets, we fit a linear mixed model with fixed effects: categorical time (0, 6, 12 months), diet, and their interaction; and with a random effect to account for the correlated observation of each participant. A Wald test of the 2-way interaction between time point and diet was used to test the primary hypothesis about diet and the change in VAT after 6 and 12 months. We quantified any preferential loss of VAT compared with SAT. However, as there is no consensus on method to accomplish this, we evaluated both the addition of SAT as a model covariate and VAT/SAT ratio (%).

To evaluate Aim 2, the effect modification by sex, we stratified the primary model by male or female, based on previous findings of sex modification in DIETFITS.¹⁰ We also evaluated effect modification by baseline insulin secretion 30 minutes after glucose consumption (INS-30) as an additional fixed effect in the primary model and testing the hypothesis using a 3-way interaction between the 12-month time point, diet, and baseline INS-30 using a Wald test.

We performed visual inspections for the model assumptions for normality of residuals and homogeneity of variance, using Q-Q plots and scatterplots, respectively. Heteroscedasticity was evaluated by plotting the residuals against predicted values. Models violating homoscedasticity were log transformed to evaluate improved fit. We provided model estimates along with 95% CI. A Satterwaite approximation for denominator degrees of freedom is used in all Wald tests. All tests were 2-sided and conducted at the 0.05 significance level.

Data availability

The data that support the findings of this study are available from the senior author, CDG, upon reasonable request.

Acknowledgements

The authors would like to acknowledge the Gardner research team from the parent DIETFITS study. We also thank Kyla Kent and Vincent Busque who contributed to the re-analysis of the DXA scans, Christopher Dant for his editorial review. We acknowledge the study participants for contributing there time and effort to this study.

Author Contributions

SF and CDG contributed to the conception; acquisition, analysis, interpretation, drafting; KMC contributed to the analysis, interpretation, drafting; MJL, MLS, and CPW contributed to the interpretation and drafting.

Materials and Correspondence: Shawna Follis

Competing Interests: No competing financial and/or non-financial interests in relation to the work to report

Landry MJ, Crimarco A, Gardner CD (2021) Benefits of Low Carbohydrate Diets: a Settled Question or Still Controversial? Curr Obes Rep 10(3):409–422. 10.1007/s13679-021-00451-z
Gardner CD, Trepanowski JF, Gobbo LCD et al (2018) Effect of low-fat VS low-carbohydrate diet on 12-month weight loss in overweight adults and the association with genotype pattern or insulin secretion the DIETFITS randomized clinical trial. JAMA - J Am Med Assoc 319(7):667–679. 10.1001/jama.2018.0245
Landry MJ, Ward CP, Cunanan KM, Fielding-Singh P, Crimarco A, Gardner CD (2024) Switching diets after 6-months does not result in renewed weight loss: a secondary analysis of a 12-month crossover randomized trial. Sci Rep 14(1):9865. 10.1038/s41598-024-60547-z
Neeland IJ, Ross R, Després JP et al (2019) Visceral and ectopic fat, atherosclerosis, and cardiometabolic disease: a position statement. Lancet Diabetes Endocrinol 7(9):715–725. 10.1016/S2213-8587(19)30084-1
Goss AM, Gower B, Soleymani T et al (2020) Effects of weight loss during a very low carbohydrate diet on specific adipose tissue depots and insulin sensitivity in older adults with obesity: a randomized clinical trial. Nutr Metab 17(1):64. 10.1186/s12986-020-00481-9
Wajchenburg BL (2000) Subcutaneous and Visceral Adipose Tissue: Their Relation to the Metabolic Syndrome. Endocr Rev 21(July):697–738. 10.1210/edrv.21.6.0415
Chaston TB, Dixon JB (2008) Factors associated with percent change in visceral versus subcutaneous abdominal fat during weight loss: findings from a systematic review. Int J Obes 32(4):619–628. 10.1038/sj.ijo.0803761
Ludwig DS, Aronne LJ, Astrup A et al (2021) The carbohydrate-insulin model: a physiological perspective on the obesity pandemic. Am J Clin Nutr 114(6):1873–1885. 10.1093/ajcn/nqab270
Garr Barry V, Stewart M, Soleymani T, Desmond RA, Goss AM, Gower BA (2021) Greater Loss of Central Adiposity from Low-Carbohydrate versus Low-Fat Diet in Middle-Aged Adults with Overweight and Obesity. Nutrients 13(2):475. 10.3390/nu13020475
Aronica L, Rigdon J, Offringa LC, Stefanick ML, Gardner CD (2021) Examining differences between overweight women and men in 12-month weight loss study comparing healthy low-carbohydrate vs. low-fat diets. Int J Obes 45(1):225–234. 10.1038/s41366-020-00708-y
Bea JW, Chen Z, Blew RM et al MRI based validation of abdominal adipose tissue measurements from DXA in postmenopausal women. J Clin Densitom Published online July 2021:S1094695021000640. 10.1016/j.jocd.2021.07.010
Schielzeth H, Dingemanse NJ, Nakagawa S et al (2020) Robustness of linear mixed-effects models to violations of distributional assumptions. Sutherland C, ed. Methods Ecol Evol. ;11(9):1141–1152. 10.1111/2041-210X.13434
Wachsmuth NB, Aberer F, Haupt S et al (2022) The Impact of a High-Carbohydrate/Low Fat vs. Low-Carbohydrate Diet on Performance and Body Composition in Physically Active Adults: A Cross-Over Controlled Trial. Nutrients 14(3):423. 10.3390/nu14030423
Allison DB, Paultre F, Goran MI, Poehlman ET, Heymsfield SB (1995) Statistical considerations regarding the use of ratios to adjust data. Int J Obes Relat Metab Disord J Int Assoc Study Obes 19(9):644–652
Frank AP, De Souza Santos R, Palmer BF, Clegg DJ (2019) Determinants of body fat distribution in humans may provide insight about obesity-related health risks. J Lipid Res 60(10):1710–1719. 10.1194/jlr.R086975
Sasakabe T, Haimoto H, Umegaki H, Wakai K (2015) Association of decrease in carbohydrate intake with reduction in abdominal fat during 3-month moderate low-carbohydrate diet among non-obese Japanese patients with type 2 diabetes. Metabolism 64(5):618–625. 10.1016/j.metabol.2015.01.012
Li X, Perelman D, Leong AK, Fragiadakis G, Gardner CD, Snyder MP (2022) Distinct factors associated with short-term and long-term weight loss induced by low-fat or low-carbohydrate diet intervention. Cell Rep Med 3(12):100870. 10.1016/j.xcrm.2022.100870

There is NO Competing Interest.

Appendix.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Effect of Low-Carbohydrate vs Low-Fat Diet Intervention on Visceral Fat in a 12-Month Randomized Controlled Trial

Status:

Version 1

Abstract

Figures

Introduction

Results

Participant characteristics

Aim 1. VAT loss in HLC vs. HLF participants at 6 and 12 months

Aim 2. Effect modification by insulin secretion status and sex

Discussion

Methods

Statistical analysis

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1