Possibly, but the best trial did not test it directly. In a 20 week feeding study of 164 adults who had just lost about 12 percent of their weight, people on a 20 percent carbohydrate diet burned roughly 209 kcal more each day than people on a 60 percent diet. Weight was held steady by design, though, so the trial measured energy burned, not who regained less.
That distinction is the whole answer. The result is strong, it is disputed on method, and it answers a narrower question than the one it usually gets quoted for.
How the trial was built
The 2018 trial in the BMJ, run by Ebbeling, Ludwig and colleagues across two US sites between 2014 and 2017, enrolled adults aged 18 to 65 with a BMI of 25 or more. Everyone first lost 12 percent of their body weight, give or take 2 percent, on a run-in diet.
They were then randomly assigned to one of three test diets for 20 weeks: high carbohydrate at 60 percent of energy (54 people), moderate at 40 percent (53), or low at 20 percent (57). Protein was held equal across all three. Energy intake was adjusted person by person to keep each participant within 2 kg of their post-loss weight. The primary outcome was total energy expenditure, measured with doubly labeled water.
The 2020 follow-up paper fills in who took part: 70.1 percent were women, and the mean run-in weight loss before randomisation was 10.5 percent. This is a group that had just done the hard part, which is exactly the population the maintenance question is about.
That design is the trial's strength and its limit at the same time. Holding weight and protein fixed isolates the effect of carbohydrate on energy burned. It also means nobody was allowed to regain.
What the numbers showed
In the intention-to-treat analysis of 162 people, total energy expenditure rose by about 52 kcal each day for every 10 point drop in the share of energy from carbohydrate. Against the high carb group, the low carb group burned 209 kcal more each day (95% CI 91 to 326). The moderate group's difference, 91 kcal, had a confidence interval running from below zero, so it was not clearly different.
Among the 120 people who held their target weight, the per-protocol gaps were larger, 131 and 278 kcal. The biggest difference showed up in the third of participants with the highest insulin secretion before weight loss, where the low versus high carb gap reached 308 kcal in the main analysis.
A 2020 follow-up in the Journal of Nutrition checked the result a second way, by how much food each person actually needed to hold their weight from weeks 10 to 20. That energy requirement was also higher on low carb, by between 181 and 246 kcal each day across the main models.
Why the result is still argued over
The 2020 paper says it plainly in its own background: the validity of doubly labeled water for comparing diets with very different macronutrient mixes has been questioned. Its authors describe their follow-up as lending "qualified support" to the method, which is a careful phrase for a reason.
Kevin Hall at the NIH, with Juen Guo and John Speakman, published a 2019 commentary asking whether low carbohydrate diets increase energy expenditure at all, and Hall discloses a running series of debates with the trial's senior author on exactly this question. The senior author, for his part, discloses royalties from books recommending a low glycemic load diet. Both disclosures are printed on the papers. Neither settles the science, and both are worth knowing before treating one number as the final word.
What the trial could not tell you
Three things sit outside this trial by construction. Weight was clamped, so it cannot say whether low carb eaters regain less in real life. Every meal was provided, so it says nothing about how easy a 20 percent carb diet is to keep up at home. And it lasted 20 weeks, while weight maintenance is measured in years.
There is also a scale effect that makes low carb look better or worse than it is. Carbohydrate is stored with water, so cutting carbs drops the scale quickly and reintroducing them brings a fast rebound that is mostly water weight, not fat. People read that bounce as proof the diet stopped working. Usually it is just glycogen refilling.
Why the Regain Radar ignores which diet you are on
OffRamp has no carb setting. It sets one nutrition number, a protein floor of 1.6 g per kilogram of goal weight, and it watches weight through a 7 day rolling average compared against the baseline you set after your loss.
The Regain Radar flags drift only when that average sits at least 2 percent above baseline on every one of 7 consecutive days, and raises an alert at 4 percent under the same rule. A two day water bounce after a pasta dinner cannot trip it, because it does not last a week. A slow real regain on any diet, low carb included, eventually will. That is the design choice: judge the pattern by the trend line it produces, not by the macronutrient label on it. The trial held protein equal for the same reason, since protein carries its own energy cost, and metabolic adaptation covers the part of maintenance that no macro split removes.
None of this is medical advice. If you have diabetes or take a medication affected by carbohydrate intake, talk to your prescriber before changing how you eat.
