what happens when we change the body’s main energy source. Usually, our cells are fuelled by glucose, a simple sugar that all other forms of sugar or starch are converted into. Glucose is highly reactive, so our bodies normally keep the amount in the blood within a narrow range to avoid damage to blood vessels and cell structures. When glucose levels rise after eating, we quickly release the hormone insulin, which tells cells to start taking up glucose, and using and storing it.

Insulin has a host of other effects, but they can be summarised as signalling to our bodies we have had an influx of calories and we need to stash them away. Crucially, insulin makes fat cells turn glucose into fat and store it. But in the absence of glucose, the body has an alternative fuel source: fat. Depending on the cell type, stored fat may be turned into fatty acids or into molecules called ketones, which can be used for energy. This normally happens a little overnight, when we go for several hours without eating any carbohydrates.

The raison d’etre of low-carbing is to minimise insulin release and be fuelled as much as possible by ketones. For most people, a shift into what is known as ketosis happens within a few days of dramatically cutting carbs. Eating very low levels of carbs is also known as a ketogenic diet.

As well as people trying to lose weight, many others have adopted low-carbing or the keto diet because they see it as a way of living healthily and prolonging lifespan. Some adherents believe entering ketosis has a range of metabolic benefits, including warding off cancer and Alzheimer’s disease, although there isn’t good evidence for this. Neurologists use very low-carb diets to induce ketosis as a treatment for certain forms of epilepsy and it is being investigated in other conditions.

Matters of the heart

What has recently granted low-carbing more legitimacy, though, is its effects on type 2 diabetes. This condition occurs when the body’s cells become less sensitive to insulin – a state known as insulin resistance – which leads to dangerous rises in blood sugar after meals. Long term, these sugar surges contribute to the many health consequences of diabetes, including nerve damage and kidney and heart disease.

The medical orthodoxy is that, because diabetes raises the risk of heart disease, it is even more important that those affected avoid saturated fats, found mainly in red meat and dairy products, because these are thought to raise blood cholesterol and lead to blocked arteries. The UK National Health Service advice for type 2 diabetes is that people should keep fat to a minimum and eat starchy foods like pasta, for instance.

Yet this ignores the fact that people with diabetes may see two benefits from low-carbing. As well as weight loss improving their insulin sensitivity, avoiding starch and sugar reduces those harmful blood sugar spikes. Remember that starch is basically long chains of sugar molecules. People with diabetes often measure their blood sugar at home, and can see for themselves that starting the day with bacon and eggs gives less of a sugar surge than toast or cereal do.

Sceptics might say that while a few days of low-carbing leads to lower blood sugar levels, it is hard to stick to this way of eating. There is mixed evidence on the issue. For instance, a review of 10 randomised trials found that low-carb diets were more effective than low-fat diets at improving blood sugar control in people with diabetes over the first year, but the differences disappeared after that. But there is evidence that for those who keep it up, the health benefits can be longer term, such as a study of 128 people with type 2 diabetes who went to low-carb counselling sessions run by David Unwin, a family doctor in Southport, UK. After an average follow-up of two years, about half had been able to stop taking all their diabetes drugs.

Because of results like these, diabetes doctors and patient support groups have started questioning the low-fat orthodoxy too. Bodies such as Diabetes UK and Diabetes Australia now say low-carbing is a valid option for weight loss. Ten years ago, that would have been unheard of. The American Diabetes Association went a step further last year and said out of all dietary strategies, low-carbing has the most supporting evidence for improving blood sugar control.

“Many are starting to question whether ‘bad cholesterol’ really is a risk to heart health”

Yet, as more and more people have adopted this way of eating, there have been renewed questions over its safety. In some cases, low-carb diets can lead to an alarming change in people’s cholesterol levels.

The idea that certain kinds of cholesterol can cause a build-up of dangerous plaques within our blood vessels is a pillar of mainstream medicine. There are several different types of cholesterol particles in the blood. One type, called high-density lipoprotein (HDL), is linked with a lower risk of heart attacks – this is sometimes known as good cholesterol. Bad cholesterol is a type called low-density lipoprotein, or LDL. Another kind of fatty particles, called triglycerides, are also thought to be harmful.

Some people on low-carb diets see their bad cholesterol levels rise significantly. Although they seem to be in the minority, the number of these “hyper-responders”, as they are coming to be termed, is unclear. Westman estimates that only a tiny fraction of people who try such diets will be hyper-responders. In one of his trials, from 2004, two people out of 59 randomised to low-carbing dropped out because their bad cholesterol levels rose.

Westman and others say they most often learn of this response in people who are slim and relatively muscular. He believes he may now be hearing about more of these cases because low-carbing is increasingly being adopted as a longer-term approach to healthy eating, not simply a short-term strategy for weight loss.

You don’t have to be slim and muscular to be a hyper-responder, though. Vipan Bhardwaj, a family doctor in Wokingham, UK, saw bad cholesterol rise in two out of 38 of his patients who began low-carbing for diabetes. “It scared the bejesus out of us,” says Bhardwaj.

What is strange about hyper-responders is that while their LDL level goes up, their other health markers tend to move in the right direction. These include their HDL, triglycerides, blood pressure and several other measures linked with glucose and insulin response. This was the case for Bhardwaj’s two hyper-responders. He got further reassurance by scanning the arteries to their heart to check for any plaques. “You see what’s really going on underneath the bonnet,” he says. “They were absolutely fine.”

Guilt by association

The fact that some people see a rise in their bad cholesterol levels on a low-carb diet while other measures improve is now leading some to question whether LDL really is a key determinant of the risk to our hearts. Much of the case against this form of cholesterol has been built on population studies done in the past few decades, which found that people with higher LDL levels were more likely to have heart attacks.

But these kinds of studies can only find correlations between blood markers and health outcomes, not prove that one leads to the other. It could be that something else is the root cause of heart disease, which increases LDL levels as a side effect.

The chief suspect for that something else is insulin resistance, says Gary Taubes, a US science journalist, who has long been a proponent of low-carbing and has just written a book called The Case for Keto.

This idea isn’t drastically at odds with our current understanding of the root causes of heart disease. Doctors already recognise that type 2 diabetes, obesity and heart disease frequently co-occur – so much so that the triad has its own name, “metabolic syndrome”. Yet, rather than heart disease being a disorder of “faulty plumbing” in which our arteries get blocked up because we consume too much fat, the real problem could be a predisposition towards insulin resistance, which in turn promotes higher insulin levels, fat storage and heart disease. “If that is true, the medical research community made a terrible mistake, and we’ve yet to fix it,” says Taubes.

Another important strand of evidence that made us believe LDL is important was the success of LDL-lowering drugs called statins, which reduce heart attack rates, according to multiple randomised trials, the most respected kind of medical study. But several other drugs that lower LDL levels don’t protect against heart attacks, and statins have many effects on the body, including dampening low-level systemic inflammation. It may be that statins actually protect the heart through their anti-inflammatory effects.

Some say it was the arrival of statins that got us so fixated on LDL levels. “Doctors only have 5 or 10 minutes with a patient. It’s convenient to write a prescription and follow the LDL,” says Bret Scher, a US cardiologist and medical director for the website, Diet Doctor, which promotes low-carb eating.

While practising physicians focus on their patients’ LDL levels, these days, cardiologists who study biomarkers of heart health debate whether LDL levels really are the most important indicator, or whether things like the ratio of total cholesterol to HDL, or that of triglycerides to HDL, would be more useful. Hyper-responders would be fine if judged by either of those metrics: generally their HDL levels rise and their triglyceride levels fall.

One trial often used to underscore the dangers of low-carb diets found that people with type 2 diabetes saw their LDL levels rise by about 10 per cent on average after a year on the regimen. However, there was improvement in most of the other 25 health markers tracked, such as weight, blood pressure and HDL. Participants’ overall risk score for heart disease fell by 12 per cent.

As an illustration of how confusing this is for the public, this particular trial is cited both as evidence against low-carbing – because LDL went up – and in favour of it, because overall risk went down. Scher acknowledges that we don’t yet know how risky it is if people on low-carb diets experience their LDL levels rising while other health markers improve. What we need are more studies that follow hyper-responders over time to see if they are developing heart disease.

In the meantime, where does the uncertainty leave the average person who wants to lose a few pounds? The emergence of hyper-responders shouldn’t stop people from trying low-carbing, says Westman. “That would be crazy, like telling someone not to take a useful drug because it sometimes has a side effect.”

On the other hand, unlike with most medicines, we don’t know how common this side effect is. Trials tend to report only average LDL changes for the whole group assigned to low-carbing.

Despite the new enthusiasm for these diets among diabetes and obesity specialists, many heart specialists and dietitians remain critical. A 2019 joint report on the prevention of heart disease from two US cardiology bodies said low-carb diets are linked with higher death rates. “The evidence is still weak about the long-term cardiovascular safety of the ketogenic diet,” says Donna Arnett at the University of Kentucky, one of the guideline authors.

“There is conflicting evidence,” says Tracy Parker, a dietitian for the British Heart Foundation. “We know saturated fat does increase your blood cholesterol.” Parker says that if people are determined to reduce their carb intake, the safest bet is to replace carbohydrates with oils from plants and fish. However, she admits that would make what is already a restrictive diet even more so, because people would have to avoid not only all starchy and sugary foods, but also meat and dairy products.

It isn’t as though low-carbing is the only way to lose weight, says Roy Taylor, a diabetes specialist at Newcastle University in the UK. Taylor has pioneered the use of meal-replacement shakes to help people quickly slim down on a low-calorie, low-fat diet, and has shown that if people with type 2 diabetes can lose about 15 kilograms this way, they can also put their disease into remission.

Unfortunately, none of the trials that have compared low-carbing with low-fat diets have lasted long enough to know which approach helps people keep weight off long term. Indeed, a recent review of many different kinds of diet – including low-carbing, low-fat and Mediterranean – concluded that most people put nearly all their lost weight back 12 months after they started anyway.

“Despite enthusiasm for these diets, many heart specialists remain critical”

Whatever works

Of course, not everyone can stick to a low-carb diet; some find they miss their bread, rice and pasta. Mike Lean at the University of Glasgow, UK, who worked with Taylor on the meal replacement diet strategy, says his obesity clinic now offers advice on both low-fat and low-carb diets. “People can use whatever they are better able to lose weight with, low-fat or low-carb,” he says. “We have found no difference in weight loss.”

The idea that different people might do better on different foods is supported by more recent research suggesting that there is no such thing as a single healthy diet that works for everyone. Instead, our individual genetics, habits and gut microbiomes may all influence how our bodies deal with the nutrients in our diet.

Yet even if the most we can say in favour of low-carb diets is that they work for weight loss and are safe for most of the population, that would still be a marked change from the previous orthodoxy that saturated fat is an inevitable route to a heart attack.

At the moment, there are more questions than answers. But even before low-carbing came along, there were growing concerns that the cholesterol theory of heart disease was on shaky ground. Now hyper-responders are making it look even wobblier. “There’s a chance that this subset of patients could upend the philosophy that LDL is the most important risk factor for heart disease,” says Scher. “I’m cautiously optimistic.”

Clare Wilson is biomedical reporter at New Scientist and author of the Health Check newsletter. Follow her @ClareWilsonMed ■