Bending the science? New keto study criticised

A new observational study has been released claiming that lean people following a keto diet with elevated LDL-C and ApoB are protected from plaque development, and therefore heart disease. But was the science bent to fit a particular narrative?  

Here we explore key criticisms of the research and its findings, and the implications for public health and scientific integrity.  

Image: The study intended to measure coronary plaque progression in lean individuals following a low carbohydrate keto diet over the course of a year (Credit: Pexels).  

What was the study?  

The observational study titled “Plaque Begets Plaque, ApoB Does Not: Longitudinal Data From the KETO-CTA Trial” was published in the journal JACC advances in April 2025. It included 100 participants already on a low carbohydrate keto diet who were classified by the authors as ‘lean mass hyper responders’ (LMHR) - lean, metabolically ‘healthy’ individuals with elevated LDL-C, HDL-C and low triglycerides. Over the course of a year, progression of plaque in their coronary arteries was measured.  

What did it find?  

According to the authors, the study found that changes in and baseline levels of ApoB were not associated with increases in noncalcified plaque volume (the amount of soft, fatty plaque buildup within arteries that is not hardened by calcium deposits), or total plaque score (the amount of calcium deposits) in the arteries. They concluded that “plaque begets plaque,” meaning that existing plaque predicts the growth of more plaque, but ApoB levels do not.  

How has it been reported on?  

The study has been used to suggest that a low-carb keto diet and high cholesterol levels may not increase the risk of heart disease as previously thought:

Key criticisms:  

There has been significant criticism of the research and its authors.  

A commentary piece condemning the research has been pre-published online, claiming the data has been misrepresented and it actually shows that participants in the study experienced rapid plaque progression. Some of the authors have also shared their views on social media.  

Here’s a breakdown of the key criticisms of the study and how it has been communicated:  

1.The authors failed to report on the registered primary outcome  

As part of good and transparent scientific protocol, researchers are required to pre-register their primary outcome before a study takes place. The primary outcome is what the study intends to measure, and pre-registering the outcome holds the authors to account. 

In this study, the pre-registered outcome was “percentage change in total non-calcified coronary plaque volume (NCPV).”  The pre-registered outcome is also required to be reported as an average value for change from baseline. 

Rather than follow this protocol, the authors instead focused on the association between Apo-B and plaque progression and only presented individual changes, instead of an average value for change.  

They also made the claim that after one year, the participants presented with ‘stable’ plaque, and one participant’s plaque even decreased, but failed to provide the data to support this.  

It has since been revealed that the NCPV of the participants actually increased by a median 18.8 mm³ over the course of a year – an increase of approximately 43% from baseline, and nearly 4 times that seen in healthy individuals.  

Critics say that the authors hid the primary outcome results, and misrepresented the data to claim that LMHR are protected from plaque progression.  

2. Lack of control in the study  

The study failed to include a low-ApoB or Low-LDL-C control group, making it hard to interpret the correlation between the risk factor (ApoB and cholesterol) and likely outcome (plaque development) in their cohort. By only including participants with high-ApoB and high cholesterol, the study lacked variation in exposure, which is essential for understanding impacts and relationships.  

Without a control group, their conclusion that changes in and baseline levels of ApoB were not associated with increases in plaque are unsurprising, as ApoB levels were high to begin with.  

3. Dishonesty in communicating the study design 

The paper described the study as a ‘trial’. Colloquially, trial can mean many things, but in science, it suggests that it was subject to a level of control. For example, the controls you might see in a randomised control trial are typically considered to result in less bias, and higher level of evidence than an observational study. But the study wasn’t a trial, and the use of the word has been labelled as misleading to readers. 

Observational studies as a category can provide rich insights into the effects of diet on related health outcome, but the reader needs to be aware of the study design in order to critically interpret results. 

4. Harm to scientific integrity – and people 

The overarching criticism of this study and the authors is that it undermines good scientific process, and misleads the community on critical area of nutrition and health research.  

It is very well established that high LDL cholesterol and ApoB increases the risk of heart disease by contributing plaque to build up in the arteries. Cardiovascular diseases (CVDs) are the leading cause of death globally, with 17.9 million people dying each year as a result. Taking this into account, inaccurate or misleading research in this space has the potential to do significant harm.  

The scientific community has a responsibility to conduct and communicate research in a manner that is evidence based, and in line with the highest standards for independence and integrity. In this case, it appears as though the authors failed to follow proper protocol, downplayed the findings and retrofitted the results to match a specific storyline.  

Takeaways:  

A new study has claimed that high cholesterol in lean individuals following a keto diet is not associated with plaque progression.  

• The study has been widely criticised, and re-examination of the data shows rapid plaque growth in these individuals.  

• The main criticisms of the research centre around the failure to report on the primary outcome, the lack of control and the misrepresentation of the study design and findings.  

• The study has the potential to mislead health professionals and patients, and undermines trust in the scientific process.  

Integrity and independence at FOODiQ Global:

As an independent organisation on a mission to positively impact global human health through food, integrity is a core value for FOODiQ Global.

We are committed to managing any potential bias or perception for bias in industry-funded research. Our mitigation strategies include pre-agreement to publish regardless of research findings, limiting funder involvement to assisting with developing the research question and providing broad topic feedback. The funders of research at FOODiQ Global do not contribute to the final methodology used, the interpretation of results, or the drafting of our manuscripts. We also pre-register the study design, and publicly share our research data.

FOODiQ Global has a longstanding commitment to implementing strategies to minimise bias and align with existing guidelines Guiding Principles: An Updated Framework for Industry Funding of Food and Nutrition Research (Larrick et al, 2022, 10.1093/jn/nxac106).