MedSysEvidence-graded reference

MedSys / Heart & cholesterol / Meal timing

Heart & cholesterol guide · Chapter 12 of 19

Meal timing — give your liver a break

What you eat is only half the picture. How long you leave your liver between meals affects hepatic triglycerides directly.

Meal timing — give your liver a break

What you eat is only half the picture. When you eat — and how long you give your liver between meals — has a direct effect on hepatic triglyceride content, insulin sensitivity, and cardiovascular risk.

Every time you eat, insulin rises and remains elevated for 3-5 hours. Insulin's job during this period is to push glucose into cells, store glycogen, and activate de novo lipogenesis — the liver's process of converting excess carbohydrate into fatty acids and packaging them as VLDL (the main carrier of triglycerides in the blood). Constant grazing keeps insulin high all day, blocking fat oxidation and continuously priming hepatic lipid synthesis.

Sugar damages two organs differently. Glucose stresses the arterial endothelium — but fructose (half of table sugar, all of agave, most of HFCS and fruit juice) bypasses glucose metabolism entirely and goes directly to the liver, where excess is converted to fat regardless of insulin levels. This is why a "low-GI" fructose-heavy food can look innocent on a glucose monitor while quietly accumulating fat in the liver — and why shortening the eating window may help clear hepatic triglycerides — though whether it does so independently of eating less is the disputed part, and is graded at the foot of this page.

During extended periods without food (after roughly 4-6 hours of nothing in), insulin drops, glucagon rises, and the liver switches modes: it depletes glycogen, mobilises fat from storage, and — critically — begins clearing the triglycerides accumulated in hepatocytes. This is the same mechanism by which time-restricted eating reverses early MASLD (formerly NAFLD). Some trials report measurable reductions in hepatic fat from shortening the daily eating window. Whether that survives once calorie intake is properly controlled is the open question — in the randomised trials that controlled for it, the improvements largely tracked weight change. See the grading at the foot of this page.

🌙
Overnight fast
12+hours minimum

e.g. last food at 7pm, first food at 7am. Below this threshold, hepatic fat clearance is impaired. Easiest single intervention — just stop eating after dinner.

⏱️
Between-meal gap
4-5hours min

Allow insulin to return to baseline before the next insulin spike. Cut out grazing and "constant nibbling" between meals.

🍽️
Eating window (TRE)
8-10hours

For stronger effect, narrow the eating window. Time-restricted eating (TRE) has been reported to reduce hepatic triglycerides, improve insulin sensitivity and lower blood lipids — but those reports come largely from uncontrolled and single-arm studies, and the randomised evidence is genuinely mixed. The summary is at the foot of this page.

🌅
Front-load the day
Earlieris better

Eating the same calories earlier in the day produces lower postprandial glucose, lower TG, and better next-day insulin sensitivity than eating them late. Chrono-nutrition research generally favours earlier eating, though mostly from short crossover studies rather than outcome trials.

🛏️
Last meal cutoff
3+hrs before bed

Late-evening eating raises overnight TG and impairs next-morning glucose tolerance. Finishing dinner at least 3 hours before sleep is associated with lower CV risk.

Fasting drinks
0calories

Water, plain tea, plain coffee, herbal teas — anything calorie-free preserves the fasted state. Milk, sugar, juice, or "just a little" of anything caloric breaks the fast.

Important caveats: time-restricted eating is not appropriate for everyone. Avoid extended fasting if you have type 1 or insulin-treated type 2 diabetes (hypoglycaemia risk), are pregnant or breastfeeding, have a history of disordered eating, are underweight, or take medications requiring food. Discuss with your GP if any of these apply. The 12-hour overnight gap is universally safe; longer windows benefit from supervision.

How good is the evidence, actually?

This chapter was the only substantive one on the site carrying no evidence grades, which was noticed when the clinical primer was reviewed — the primer asserted that the metabolic lipid pattern responds to shortening the eating window, and pointed here, and here graded nothing. A site built on evidence grading cannot have a page that quietly opts out. So:

ClaimGradeWhy
The overnight fast lowers insulin and shifts the liver toward fat oxidation Strong Basic physiology, directly measurable, not in dispute.
Constant grazing keeps insulin elevated and suppresses that switch Moderate Mechanistically sound and consistent with feeding studies; the long-term outcome data are thinner than the mechanism suggests.
Time-restricted eating improves metabolic markers beyond the effect of eating less Contested This is the disputed claim, and it is the one most often stated as settled. See below.
TRE is an easier way for some people to eat less Moderate A rule about when is easier to follow than a rule about how much, for some people and not others. This is probably the case for it.
Eating late at night is worse than the same food earlier Emerging Circadian mechanism is real and there are short crossover studies; hard outcome data are absent.
Why the third row is Contested rather than Moderate. Early uncontrolled and single-arm studies of TRE were striking, and much of the popular enthusiasm dates from them. The randomised picture is less flattering: TREAT (2020) found no significant benefit against a control eating pattern, and a 2022 randomised trial in the New England Journal of Medicine found that adding time restriction to calorie restriction produced no additional benefit — with the improvements in both arms tracking weight change rather than timing. The other side, stated fairly: several controlled feeding studies in which calories were matched by design — early time-restricted feeding work in particular — have reported improvements in insulin sensitivity and blood pressure without weight loss, which is the strongest affirmative evidence that timing does something of its own. They are small and short, which is why this is Contested rather than settled in either direction. That does not make TRE useless. It makes the claim that timing has an effect independent of intake a genuinely open question, and it means anyone promising metabolic benefits without weight loss is ahead of the evidence.

What survives all of that is modest and worth keeping: a long overnight gap is free, physiologically sensible and easy; not grazing continuously is a reasonable default; and if a shorter eating window helps you eat less, that is a good enough reason on its own without needing a special mechanism. What does not survive is treating the eating window as a metabolic intervention in its own right.