Two people start the same low-carbohydrate plan in January. By summer, one has an LDL cholesterol reading that alarms their doctor, and the other has a number barely different from where they started. Both describe the experience online, both are telling the truth, and neither explains the other.
Research presented this week offers part of the reason, and with it a practical step that requires no genetic testing at all.
Alexa Barad, a postdoctoral scholar at the Stanford University School of Medicine, and colleagues analyzed genetic and dietary data from DIETFITS, a randomized trial in which more than 600 adults followed either a healthy low-carbohydrate or a healthy low-fat diet for a year. Among the 431 participants with genetic data available, those carrying a genetic predisposition toward higher LDL cholesterol were the ones most likely to see it climb on the low-carb arm. Participants on the low-fat diet did not show the same pattern.
The mechanism the analysis points to is sensitivity to saturated fat. People at higher genetic risk had the largest LDL increases when they ate more of it.
Why This Explains the Argument Rather Than Settling It
Low-carbohydrate eating has an unusually polarized evidence base, and some of that polarization is an artifact of averaging. Trials report a mean change, and the mean hides a wide spread.
Barad described the consequence plainly, saying the findings suggest "some individuals may be more sensitive to the LDL cholesterol-raising effects of saturated fat" in the context of a low-carbohydrate diet because of their genetic background. Two people making opposite claims about what keto did to their cholesterol can therefore both be reporting accurately.
Body weight appears to be another major source of variation. A meta-analysis of 41 randomized trials published in The American Journal of Clinical Nutrition found that baseline body mass index had a strong inverse relationship with LDL change on low-carbohydrate diets. In trials where participants averaged a BMI under 25, LDL cholesterol rose by about 41 mg/dL. In trials averaging a BMI of 25 to under 35, it did not change measurably, and in trials averaging a BMI of 35 or higher it fell by about 7 mg/dL.
Put those findings together, and the picture is that leaner people with less favorable genetics are the group most likely to see a large increase, and also the group least likely to be expecting one.
What a Lipid Panel Actually Measures
A standard lipid panel is an inexpensive, widely available blood test reporting total cholesterol, LDL cholesterol, HDL cholesterol, and triglycerides. Most laboratories now report non-HDL cholesterol as well, and some add apolipoprotein B, which counts atherogenic particles rather than the cholesterol they carry.
The value of testing before and after a major dietary change is that it replaces a population average with your own number. A trial can say what happened to 431 people on average. A pair of your own results says what happened to you.
Timing matters for interpretation. Lipid changes from a substantial shift in dietary fat generally appear within several weeks, so clinicians commonly look at a repeat panel a couple of months after starting, though the right interval is a conversation with a clinician rather than a fixed rule. Fasting is no longer required for most routine panels, but keeping conditions consistent between the two draws makes them easier to compare.
None of this is a reason to start or stop a diet on your own. It is a reason to have a number to discuss.
Who Should Be Most Careful
The people with the most at stake are those who already have elevated LDL cholesterol, a personal or family history of early heart disease, familial hypercholesterolemia, diabetes, or existing cardiovascular disease. For them, a diet that raises LDL substantially is not a neutral experiment, and the decision belongs with a clinician before the diet starts.
Lean, weight-stable adults deserve particular attention, because the evidence suggests they are most likely to see a large rise and least likely to be watching for one. Published case series have documented LDL levels several times baseline in individuals following very-low-carbohydrate diets, often in people who considered themselves healthy.
Anyone taking lipid-lowering medication should not adjust it around a dietary change without medical guidance. Pregnant people, adolescents, and anyone managing kidney disease should treat carbohydrate restriction as a medical decision rather than a lifestyle one.
What the Research Does and Does Not Show
The limitations are real, and the researchers state them.
This is a conference abstract that has not completed peer review, and it is a secondary analysis of a trial designed to answer a different question. It measured change in LDL cholesterol from baseline to six months, not heart attacks or strokes, so it does not establish what these lipid changes mean for long-term cardiovascular outcomes in this population.
The polygenic score approach, which combines thousands of genetic variants rather than examining single genes, was developed largely in populations of European ancestry and may perform differently in others. The researchers say further work is needed to confirm the findings apply broadly.
Nothing here supports ordering a genetic test. Barad framed polygenic scores as a future clinical possibility rather than a current tool, and noted that the immediate lesson holds without any genetic testing. Responses vary, so monitor rather than assume.
What Readers Can Reasonably Do
Anyone considering a ketogenic or low-carbohydrate plan can ask a clinician for a baseline lipid panel before starting and a repeat afterward, and can raise family history of early heart disease in the same conversation.
Within a low-carbohydrate pattern, the composition of the fat is modifiable even when the carbohydrate target is not. General dietary guidance recommends keeping saturated fat below 10 percent of daily calories. Emphasizing nuts, seeds, olive oil, avocado and fish over butter, beef tallow, fatty cuts of meat and processed meats is the adjustment researchers suggest may reduce the risk of an adverse cholesterol response.
What is not warranted is abandoning a diet that is working because of a headline, or continuing one without ever checking, which is the more common error.
What Happens Next
Peer-reviewed publication would allow independent review of the polygenic scoring and the dietary assessment. Larger studies in more diverse populations are the step the researchers identify as necessary before polygenic scores could guide dietary counseling in practice.
The newest finding is that genetic background helps explain who sees LDL cholesterol rise on a low-carbohydrate diet. The people most affected are lean adults with a genetic predisposition toward higher cholesterol. The most reasonable step is a lipid panel before and after, discussed with a clinician. The central uncertainty is what these lipid changes mean for cardiovascular events over a lifetime.
Frequently Asked Questions
What did the study find? Among 431 participants in the DIETFITS trial, those with a genetic predisposition toward higher LDL cholesterol were most likely to see it rise on a low-carbohydrate diet, driven by greater sensitivity to saturated fat.
Should I get genetic testing before trying keto? No. The researchers describe polygenic scores as a possible future clinical tool, not a current one. The actionable step is a lipid panel.
When should I recheck my cholesterol? Clinicians commonly look at a repeat panel a couple of months after a major dietary change, but the right interval depends on your situation and should be set with your clinician.
Does everyone's cholesterol rise on low-carb diets? No. Trial data show leaner people tend to see large increases while people with higher BMI often see no change or a decrease.
What does a lipid panel measure? Total cholesterol, LDL, HDL and triglycerides, usually with non-HDL cholesterol. Some panels add apolipoprotein B.
Can I change the diet to reduce the risk? Researchers suggest emphasizing nuts, seeds, olive oil, avocado and fish over butter, fatty meats and processed meats, and keeping saturated fat under 10 percent of calories.
Who should talk to a doctor first? Anyone with high cholesterol, diabetes, existing heart disease, familial hypercholesterolemia, or a family history of early heart disease.
