A network meta-analysis published online June 24, 2026 in Clinical Nutrition ESPEN does what single fasting trials cannot: it ranks the fasting styles against each other. The team pooled 19 randomized controlled trials with 1,149 adults with MASLD and compared time-restricted feeding, alternate-day fasting, and the 5:2 diet head to head, then matched each one to the outcome it moves best. The result is a practical menu instead of a one-size-fits-all claim.
Time-restricted feeding, eating all daily food within a set window, came out on top for the liver itself. It reduced hepatic steatosis by about 23.3 dB/m, the largest effect of any regimen, and the 5:2 diet, two lower-calorie days a week, came second. Both also significantly lowered the liver enzymes ALT and AST, which means the benefit showed up in the bloodwork a clinician actually orders, not only in research imaging.
The metabolic wins spread across the styles. Alternate-day fasting produced the largest reductions in body weight, waist circumference, and body fat, and it was the only regimen to significantly improve HOMA-IR, the standard marker of insulin resistance. The 5:2 diet added a modest but significant drop in LDL cholesterol. A reader who fasts is not choosing one outcome; the data say the style decides which outcome improves most.
The analysis also drew a clean line on what fasting does not do. None of the regimens significantly improved liver stiffness, the proxy for fibrosis, which is the honest limit of the approach. Fasting targets the fat and the metabolic engine, not the scar, and a person with suspected advanced fibrosis needs a clinician to measure the liver directly rather than assuming an eating pattern alone will reverse it.
Why does any of this work? Fasting concentrates eating into a window or onto fewer days, which typically trims total energy, shifts when the liver receives substrate, and improves insulin sensitivity across the day. The mechanisms overlap with the circadian biology covered earlier this week: the liver handles food better in some hours than others, and fasting is the bluntest way to put that rhythm to work without counting every calorie.
The practical translation for a reader is a choice among three tolerable patterns, not a single command. Someone whose main goal is liver fat can try an eight-to-ten-hour eating window; someone chasing weight and insulin can try alternate-day style fasting; someone watching cholesterol can try the 5:2 rhythm. The meta-analysis hands each goal its best match, which is more useful than generic advice to eat less.
The adherence story matters as much as the numbers. Continuous calorie restriction works but is hard to keep, which is why the authors frame fasting as a more sustainable alternative. A pattern a person can actually maintain for months beats a stricter one abandoned in three weeks, and the trial evidence inside the meta-analysis reflects real people living with the regimens, not laboratory perfection.
The limits of a network meta-analysis deserve a plain statement. The 19 trials are heterogeneous in window length, diet quality, and follow-up, and the network compares them indirectly, which adds uncertainty even with confidence intervals. The authors are careful to call for more direct comparisons, and a reader should treat the ranking as directional guidance rather than a stopwatch-grade prescription.
The food inside the window still rules. A fasting window filled with fried food and sugar will not rescue the liver, and a reader who uses fasting as a license to eat poorly inside the hours is missing the point the trials make. The friendly plate, vegetables, legumes, whole grains, fish, nuts, and unsaturated fat, is the engine, and fasting is the schedule that lets the engine run in its best hours.
The interplay with diabetes medication needs a red flag. Fasting changes glucose and insulin handling, and people on insulin or sulfonylureas can develop hypoglycemia on restricted days. The analysis is about people with MASLD, not a green light to rearrange medication, and any reader with diabetes should review a fasting plan with the prescribing clinician before the first fasted day.
None of this replaces the clinician or the scan. The meta-analysis pools trial data on steatosis and metabolic markers, not a treatment plan for a person with confirmed fibrosis or cirrhosis, and the stiffness finding is exactly the reminder that fasting has a ceiling. The eating window is the daily lever a reader controls; the elastography, the biopsy, and the prescription belong to the care team.
The three-way ranking also gives readers a way to start small. Time-restricted feeding is the gentlest entry, a fixed window that does not require fasting days or calorie counting, and it is the pattern with the strongest liver-fat signal. A reader new to fasting can hold an eight-hour window for a month and let the evidence, not the hype, decide whether the style suits their life.
The analysis quietly supports what this site has said about timing all week: when the food arrives matters, and the liver has preferred hours. Fasting is the blunt tool that enforces those hours, and the meta-analysis gives it a fair hearing with real trial numbers. The plate stays the foundation, and the window is the frame that makes the plate work harder, which is a hopeful, actionable message for a reader tired of calorie counting.
A reader should also expect the fasting field to keep moving. Direct head-to-head trials are coming, and the ranking may shift as windows and controls standardize. Until then, the sensible read is the one the analysis itself offers: pick the pattern that fits the goal and the life, keep the food inside the window friendly, and let the liver enzymes and the scale tell the truth about whether it is working.
Which fasting style worked best for liver fat?
Time-restricted feeding, eating within a daily window, reduced hepatic steatosis by about 23.3 dB/m, the largest effect of any regimen in the network. The 5:2 diet came second, and both significantly lowered ALT and AST.
Did fasting help with weight and insulin?
Alternate-day fasting produced the greatest drops in body weight, waist circumference, and body fat, and it was the only regimen to significantly improve HOMA-IR, a marker of insulin resistance. The 5:2 diet also gave a modest LDL cholesterol reduction.
Does fasting slow liver scarring?
None of the fasting regimens significantly improved liver stiffness in the analysis, which is an honest limit. Fasting appears to target steatosis and metabolic risk rather than established fibrosis, and anyone with advanced disease needs clinical care.