Othercited by the Tayebat archivefull text checked★ key study A-SOD-07 in-depth review
Fructose vs glucose drinks for 10 weeks
Consuming fructose-sweetened, not glucose-sweetened, beverages increases visceral adiposity and lipids and decreases insulin sensitivity in overweight/obese humans.
Overview
Stanhope KL et al.. J Clin Invest 2009 · PMID 19381015 · DOI 10.1172/JCI37385
At very high doses, fructose drinks increased belly fat and worsened blood fats more than glucose drinks.
Does the study support the specific claim? Partly · Methodological quality: low
Plain-language summary
Study explained simply
- What did the researchers want to find out?
Do fructose and glucose differ in metabolic effects during sustained consumption?
- How was the study done?
Controlled parallel-arm intervention; beverages providing 25% of energy for 10 weeks.
- How many people, animals or samples were studied?
32 completers (17 fructose, 15 glucose) of 39 enrolled overweight/obese adults aged 40-72
- What was actually found?
Similar weight gain in both groups; visceral adipose volume increased only with fructose; fasting TG +~10% with glucose but not fructose; DNL, postprandial TG, apoB, LDL, small dense LDL, oxidised LDL increased with fructose; fasting glucose and insulin increased and insulin sensitivity decreased with fructose.
- How large was the observed effect?
Visceral fat +14.0% (fructose) vs +3.2% (glucose); insulin sensitivity -17.3% vs +1.1%; apoB +27.2% vs +3.0%
- What do the results mean in everyday life?
Very high fructose intake can harm metabolism.
- What are the study's limitations?
Very high dose (25% of energy); small, matched (not clearly randomised) groups; completers only; many outcomes; liver fat not measured
- What does it say about Elawady's specific claim?
Mechanistic support for C-SOD-04.
- Do the results apply to Germany and Europe?
High (physiology).
- What can explicitly not be concluded from this study?
That normal soda intake has these effects.
Scientific in-depth review
1. Full reference
- Authors
- Stanhope KL, Schwarz JM, Keim NL, Griffen SC, Bremer AA, Graham JL, Hatcher B, Cox CL, Dyachenko A, Zhang W, McGahan JP, Seibert A, Krauss RM, Chiu S, Schaefer EJ, Ai M, Otokozawa S, Nakajima K, Nakano T, Beysen C, Hellerstein MK, Berglund L, Havel PJ
- Journal
- J Clin Invest
- Year
- 2009
- Identifier
- PMID 19381015 DOI 10.1172/JCI37385 · PMC2673878
- Verification
- verified – PubMed E-utilities efetch, 2026-10-25
- Retracted
- no
- Corrections
- none
- Extent checked
- full text
- Conflicts of interest and funding
NIH grants (incl. HL-075675, UL1 RR024146) and American Diabetes Association support (partly truncated in file). Several authors report industry ties: Pfizer shares (Berglund), KineMed stock/employment and consulting (Beysen, Hellerstein), Bristol-Myers Squibb income (Griffen), Novo Nordisk research contract (Havel), Merck/Sanofi/Isis/Metabolex grants or consulting (Krauss), Denka Seiken/Otsuka/Kyowa Medex fellowships or consulting.
2. Research question
Do fructose and glucose differ in metabolic effects during sustained consumption?
3. Study design and population
- Study design
- Other – Double-blinded parallel-arm controlled feeding intervention with matched allocation (groups matched for sex, BMI, fasting TG and insulin; random allocation not described): 2-week inpatient baseline, 8-week outpatient ad libitum phase, 2-week inpatient energy-balanced phase
- Population
Overweight/obese adults aged 40-72 (BMI 25-35), postmenopausal women and men; without diabetes, renal/hepatic disease, hypertension, TG >400 mg/dl; non-smokers; habitual intake <=1 sugar-sweetened beverage/day
- Country
USA
- Period
Not stated in full text (published May 2009); 12-week protocol per participant
- Sample size
39 enrolled; 32 completed (17 fructose, 15 glucose); 7 dropouts (4 fructose, 3 glucose); DNL measured in a subset (18 successful tests: 8 glucose, 10 fructose); CT n=14 glucose
- Inclusion and exclusion criteria
Age 40-72, BMI 25-35, stable weight; excluded diabetes, renal or hepatic disease, fasting TG >400 mg/dl, hypertension >140/90, bariatric surgery, smoking, vigorous exercise >3.5 h/week, thyroid/lipid/glucose/BP/antidepressant/weight-loss medication, >1 SSB/day or >2 alcoholic drinks/day
4. Methods and endpoints
- Exposure / intervention
Fructose-sweetened beverages providing 25% of energy requirements (617 +- 24 kcal/day) for 10 weeks
- Comparator
Glucose-sweetened beverages at 25% of energy requirements
- Endpoints
Body weight and composition (DXA, CT abdominal fat), adipose gene expression, 24-h TG profiles, lipoproteins (apoB, LDL, sdLDL, oxLDL, RLP), hepatic fractional DNL (stable isotopes), postheparin LPL, OGTT and insulin sensitivity index
- Statistical methods
Repeated-measures mixed models (time, sex, metabolic syndrome risk factors; previous-day energy as covariable); GLM ANOVA for 2-time-point outcomes; paired t tests within groups; Tukey post tests; completers only
5. Actual results
- Results
Both groups gained weight and fat (total body fat +3.2% glucose, +2.8% fructose). Intra-abdominal (visceral) fat +3.2 +- 4.4% (glucose, n.s.) vs +14.0 +- 5.5% (fructose, P<0.01 within-group); greater in men (+18.1%) than women (-0.6%). Fasting TG +9.7% (glucose, P<0.05) vs +3.9% (fructose, n.s.). 23-h TG AUC -32.0 vs +99.2 (fructose P<0.001). Fructose: fasting LDL-C +13.9%, apoB +27.2%, sdLDL +44.9%, oxLDL +12.8%, postprandial DNL 11.4% -> 16.9% (P=0.021; 16-h DNL AUC between-sugar P=0.016), postprandial LPL activity difference between sugars P=0.041. Fasting glucose -1.4% (glucose) vs +5.3% (fructose); fasting insulin +10.2% with fructose; insulin sensitivity index +1.1% vs -17.3% (fructose, P<0.01). Women had greater decreases in insulin sensitivity with both sugars. Blood pressure unchanged. Liver fat not measured.
- Absolute risks
not applicable
- Authors' conclusion
Dietary fructose specifically increases DNL, promotes dyslipidaemia, decreases insulin sensitivity and increases visceral adiposity in older overweight/obese adults; dose-response studies are needed; the 25% energy dose exceeds mean US added-sugar intake (15.8%), and whether sucrose/HFCS have similar effects is uncertain.
6. Strengths
- Tightly controlled inpatient metabolic phases with energy-balanced diets
- Blinded beverages with riboflavin compliance biomarker
- Detailed gold-standard metabolic phenotyping (stable isotope DNL, CT, 24-h profiles)
7. Weaknesses
- Small sample (32 completers); DNL subset only 18
- Allocation by matching; randomisation not described
- Many outcomes and within-group paired tests; several 'differential' effects inferred from within-group significance rather than formal between-group tests
- Completers-only analysis (7 dropouts)
- Very high dose (25% of energy as pure fructose) vs mixed sugars in real soft drinks (authors)
- Weight and fat gain in outpatient phase confound insulin-sensitivity findings (authors)
- Older overweight/obese adults only
- Liver fat not measured
- Authors' industry ties
8. Risk of bias
- Orienting tool
- RoB 2 (applied cautiously; allocation by matching, randomisation not described - ROBINS-I would also apply)
- Type of assessment
- structured assessment based on the full text by a single reviewer (no formal double review)
- Full formal assessment
- no
- Overall risk of bias
- high
- Rationale
Changed from some_concerns: full text shows matched (not described as randomised) allocation, small completers-only sample and many within-group comparisons. Structured single-reviewer judgement from full text; a formal assessment would require two independent reviewers and the protocol.
| Domain | Judgement | Rationale |
|---|---|---|
| D1 randomisation process | high | Groups 'matched' for sex, BMI, TG and insulin; no description of random sequence generation or concealment. |
| D2 deviations from intended interventions | low | Double-blind beverages, riboflavin compliance marker similar between groups. |
| D3 missing outcome data | some concerns | 7/39 dropouts, completers only; DNL available in subset only due to budget and catheter failures. |
| D4 measurement of the outcome | low | Objective laboratory and imaging outcomes with blinded technicians. |
| D5 selection of the reported result | high | Very many outcomes and time points; conclusions of 'specific' effects often based on within-group significance; no multiplicity correction; no pre-registration reported. |
Full-text review (25 October 2026): Added sample (39 enrolled/32 completers), eligibility, matched allocation, dose (617 kcal/day), all key percentage changes (VAT +14.0% vs +3.2%, insulin sensitivity -17.3%, apoB +27.2%, fasting TG +9.7% with glucose), DNL subset size, sex differences, author industry COI; design_category changed from rct to other (non-randomised matched intervention); bias some_concerns -> high.
9. Assessment of the statistical interpretation
Physiologically important mechanistic signals at a supra-typical dose of pure fructose; between-group contrasts are small-sample and partly inferred from within-group changes, so effect sizes are imprecise. Glucose also raised fasting TG and both sugars caused weight and fat gain.
10. Relevance to the specific claim
Mechanistic support for C-SOD-04.
11. Independent contrary findings
I-SOD-04: no effect in isocaloric exchange.
12. Transferability to Germany and Europe
High (physiology).
13. Overall methodological judgement
Important mechanistic study supporting fructose-specific lipogenic and dyslipidaemic effects at very high intake; not direct evidence for usual soft-drink consumption, and liver fat was not measured.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | low |
| Risk of bias | high |
| Precision of results | low |
| Directness | indirect |
| Reproducibility | partly replicated |
| Consistency with the overall evidence | mixed |
| Regional transferability | high |
| Adequacy of the interpretation (by the archive) | partly correct but over-interpreted |
14. Open questions
Effects at realistic doses and with sucrose/HFCS; replication with randomised designs; liver fat outcomes.
15. Linked fact checks
- Fructose in soft drinks 'overwhelms the liver': compared with the same calories from glucose, it specifically increases visceral fat, LDL cholesterol and insulin resistance, and soft drinks promote fatty liver. – Studies supporting the claim
- 'Insulin resistance' is a wrong term; there is no such thing as insulin resistance. – Studies contradicting or limiting the claim
Use by the Tayebat archive
Note: according to the archive, the studies were compiled by the archive operators. Whether Dr. Elawady himself cited this study is not established.
- What the archive says about the study
Controlled trial: 10 weeks of fructose-sweetened beverages increased visceral fat, raised cholesterol and reduced insulin sensitivity — the same calories from glucose did not.
- A. What was actually studied?
Fructose vs glucose at 25% of energy.
- B. What did the authors find?
Fructose-specific adverse changes; glucose raised fasting TG.
- C. What conclusions do the authors themselves draw?
Fructose specifically harmful in this setting.
- D. What does the archive derive from it?
Generalises to soda consumption and says the glucose group 'suffered none of this'.
- E. Is this derivation scientifically justified?
Partly overinterpreted: glucose also raised fasting triglycerides (+9.7%), both groups gained weight and fat, the dose (25% of energy as pure fructose, about 617 kcal/day) far exceeds typical soda intake, allocation was by matching in 32 completers, and liver fat was not measured, so 'overwhelms the liver/fatty liver' is not shown by this study.
- Adequacy of the interpretation (by the archive)
- partly correct but over-interpreted