The Study That Exists: The Norwegian 4x4 and Coronary Plaque

“The Norwegian 4x4 interval training protocol (four 4-minute bouts at 85–95% of maximal heart rate separated by 3 minutes of active recovery) produces measurable regression of coronary atherosclerotic plaque and an increase in VO2 max.”

Where it circulates

The Proof (Simon Hill)

Can Supplements Reverse Your Atherosclerosis? Testing 8 Options Against the Evidence

Jun 29, 2026 1:03:16 89,346 views on Sep 20, 2026

@c6dd68 3mo ago

A commenter proposing exercise as another route besides supplements and drugs: the Norwegian 4x4 interval protocol, they write, has been shown in a study to measurably shrink coronary plaque, as well as raising VO2 max.

9 likes 2 replies

Nothing in this episode supports the claim. The comment originated here, but the claim let go of its source – that is a finding, not a gap.

The adjudicated claim is the commenter's, not the host's.

Peter Attia MD

Building strength and muscle mass: optimize training & nutrition for longevity (AMA #71 rebroadcast)

Jul 6, 2026 1:57:50 45,786 views on Sep 20, 2026

@357a30 3mo ago

A commenter arguing against maximising VO2 max: too much exercise, they write, drives inflammation and plaque, and balance matters more than volume.

2 likes 2 replies

Nothing in this episode supports the claim. The comment originated here, but the claim let go of its source – that is a finding, not a gap.

The adjudicated claim is the commenter's, not the host's.

Huberman Lab

Build Muscle, Great Posture & Resilience to Injury | Jeff Cavaliere

May 25, 2026 2:16:50 232,391 views on Sep 20, 2026

@50989f 4mo ago

A commenter recommending short, fast intervals, on a track, a bike or in a pool, for conditioning, noting that speed work raises VO2 max, which is associated with longevity.

0 likes 0 replies

Nothing in this episode supports the claim. The comment originated here, but the claim let go of its source – that is a finding, not a gap.

The adjudicated claim is the commenter's, not the host's.

Huberman Lab

How Mitochondria Control Your Metabolism | Dr. Jared Rutter

Sep 7, 2026 2:03:47 333,675 views on Sep 20, 2026

@27b815 1mo ago

A commenter marvelling at how heart muscle can repair and renew itself, citing an acquaintance who follows a demanding routine of zone-two walking, interval sprints and long fasts.

1 like 0 replies

Nothing in this episode supports the claim. The comment originated here, but the claim let go of its source – that is a finding, not a gap.

The adjudicated claim is the commenter's, not the host's.

A comment under a podcast episode about reversing atherosclerosis made a claim precise enough to test. It named a training protocol, the Norwegian 4x4: four minutes hard, three minutes easy, four times through. It named two outcomes: a higher VO2 max, and a measurable shrinking of coronary plaque. And it said the evidence existed, in a study.

We looked. The fitness outcome has been measured in randomised trials that used this exact prescription, and it holds. The plaque outcome has been measured too: once, directly, against standard care, in a randomised trial of high-intensity intervals in Norway. That trial says less than the comment says. Its main question came back with no difference. A second measure, plaque volume, fell slightly more with training.

So the comment was pointing at something real. The rest of this page is about the size of it, and what surrounds it.

What everyone says

The claim, whole or in part, reached us as 5 occurrences across 3 channels and 4 videos, posted between 01/06/26 and 08/09/26. The count is fixed as of our collection round of 18/09/26. Three channels is where the question appears, not a count of people: the collection does not store authors, by design, so distinct commenters are not measured.

The four episodes are about different things: eight supplements tested against the evidence on atherosclerosis; muscle mass and strength; mitochondria and metabolism; and, in an interview about muscle, posture and injury, conditioning in general terms: cardio, intervals, zone two, a skipping rope, a bike.

We read the automatic transcripts of all four, end to end, looking for the claim. It is in none of them. None names the Norwegian protocol, none mentions plaque regression from training, and none attaches a VO2 max figure to this method.

In most claims we adjudicate, the pattern runs the other way. A host says something with a condition attached, and the condition falls off on the way down to the comments. Here the recordings said nothing on the point, and the audience arrived with something more specific than anything on them: one protocol, two outcomes, and a statement that a study existed. We record that as a finding.

The argument underneath

The comment was not alone. Under a different episode, on a different channel, another comment argues the reverse: that too much exercise, in pursuit of an ever higher VO2 max, drives the inflammation that puts plaque into the artery. Under the same episode as the first comment, a long-time endurance trainee writes that their own scan shows plaque anyway, and asks whether others are in the same position. That comment drew more replies than any other in the group.

Three comments, three starting points, one question underneath: whether hard cardio clears the artery or damages it. The four episodes were not having that argument. The comments underneath them were. Across the 166 comment texts gathered for this dossier, nine arguments split close to even: four for the claim, five against. Most of the five say plaque cannot be exercised away, rather than that exercise puts it there.

What the evidence shows

98 published sources. The claim carries two assertions inside one sentence, and they separate the moment they are tested. Three questions, answered separately.

Direction of effect

Partial. Evidence: high.

The fitness half holds, with the exact prescription. In a 12-week supervised trial in type 2 diabetes, 4x4 intervals at 85–95% of maximum heart rate raised VO2 max from 25.6 to 30.9 mL/kg/min, about a fifth, and more than moderate training did. Pooled across 23 randomised trials and 1,730 patients with coronary disease, exercise training raised peak oxygen uptake by 2.86 mL/kg/min (95% CI 2.00–3.73), with high-intensity intervals among the largest effects, 3.60 mL/kg/min. Most of these trials report peak oxygen uptake, the measure closest to the VO2 max the comment refers to.

Three qualifications come from the supportive trials themselves.

First, the response does not repeat inside the same person. When 14 healthy young men completed two identical four-week blocks of 4x4 training three months apart, the group gained both times, but the individual response in one block barely predicted the other (intraclass correlation 0.369, where 1.0 would mean the same response both times). In five pooled sprint-interval studies of 63 young adults, a different format, 22% of participants did not improve beyond measurement error.

Second, time. In 93 patients with coronary disease confirmed on angiography, the 4x4 beat 40 minutes of continuous training after four weeks (10% against 4%, P = 0.02). At 12 months the advantage was gone (10% against 7%, P = 0.30), once training moved home and only about half of the interval group stayed adherent. A protocol that works under supervision, over a few weeks, in group means, is not the same claim as a protocol that works.

Third, intensity. In a three-arm trial of 135 patients after coronary revascularisation, Nordic walking at the same moderate heart-rate target as the continuous arm improved six-minute walking distance more than the 4x4 did: 77.2 m against 51.4 m (interaction P = 0.042). Harder was not better in that trial. Head-to-head trials in coronary patients disagree: in SAINTEX-CAD, 200 patients gained equally with intervals and continuous training (22.7% against 20.3%); in a 382-patient multicentre trial, short intervals gained 1.04 mL/kg/min more (P = 0.002).

The plaque half is a different kind of problem. None of the trials in the dossier that used the 4x4 prescription imaged a coronary artery. In CENIT (NCT02494947), a Norwegian trial, 60 patients with stable coronary disease were randomised after angioplasty to six months of supervised high-intensity intervals at 85–95% of peak heart rate, twice a week, or to standard preventive care, and had their coronary arteries imaged from inside, before and after. The published summaries do not say whether the intervals were four minutes long, so we do not call it a 4x4 trial.

The trial’s primary question was the fat inside the plaque, measured by near-infrared spectroscopy in the 49 patients with enough of it to analyse. The estimated fall in plaque fat was about 47 points in each group; the difference between them was −1.2 (95% CI −65.8 to 63.4, P = 0.97).

A second paper from the same trial reported plaque volume by intravascular ultrasound, and there the training group came out ahead. Percent atheroma volume, the proportion of the vessel’s volume occupied by plaque, fell by 1.2 points with training and did not change in the control group: a between-group difference of −1.4 points (95% CI −2.7 to −0.1, P = 0.036). Total atheroma volume, normalised for segment length, differed by −12.0 mm³ (95% CI −19.9 to −4.2, P = 0.003). The authors conclude that the training reduced atheroma volume.

That result is real, and it is the closest thing in the record to what the comment described. Four things limit it. It is not the trial’s primary endpoint, and the primary endpoint returned no difference. It comes from 60 patients. Both arms followed standard preventive care, so it shows training added to that care, not instead of it. And we found no second trial of interval training that repeats it against standard care. A JACC state-of-the-art review adds that plaque regression on imaging has not yet been shown to mean fewer cardiovascular events.

Other trials frame it. In 36 patients after stenting, intervals at about 90% of peak heart rate and continuous training at about 70% both reduced necrotic core, and no imaging measure differed between them. In 23 patients with diabetes and coronary disease, six months of a programme centred on exercise training improved coronary artery function and left plaque burden unchanged. In an older randomised trial of 62 coronary patients, a year of a regular exercise programme was followed by regression on angiography in 28% of patients against 6% of controls; the patients whose lesions regressed averaged about five to six hours of exercise a week. A 2026 scoping review of five controlled trials, 217 patients, reports lower plaque volume with exercise-based rehabilitation, and itself flags confounding from lipid-lowering drugs as unresolved.

The counter-claim holds only in part. A systematic review of 33 studies (PROSPERO CRD42024607693) found that 9 of 12 comparative studies above 450 minutes of exercise a week showed higher, not lower, calcium scores in the heavy trainers. The same review reports that, where plaque composition was examined, it was mostly the more benign, calcified kind, and that most studies tracking clinical events found no excess. A joint consensus statement of the EAPC and the ACC reports that masters athletes may have more coronary atherosclerosis than less active peers, and calls for outcome studies. Among 198 male masters endurance athletes at low predicted risk, 71% had detectable coronary calcium, 12% a narrowing above 50% and 13% high-risk plaque. Two narrative reviews in the dossier note that the benign reading of athletes’ plaque has been questioned.

Among heavy trainers, most studies that tracked events have not found that a higher calcium score comes with more heart attacks. That is a finding about this group, not about what a high score means for anyone else: calcium in general, the same review notes, is a robust predictor of cardiovascular events. Reading heavy trainers’ calcium as their risk is the same shortcut, in reverse, as reading a shrinking scan as fewer heart attacks.

Volume alone did not speed the calcium up. In 8,771 apparently healthy adults scanned twice, a mean of 7.8 years apart, annual progression differed by 0.0% per 500 MET-minutes of activity a week (95% CI −0.1% to 0.1% in men, −0.4% to 0.5% in women). The cohort measured activity in general, and its outcome was calcium, not plaque volume.

Thresholds

Fails. Evidence: moderate.

The numbers in the protocol read like physiology. They are a prescription.

There is no common definition of high-intensity interval training. A 2023 perspective paper says so directly. A 2015 crossover study of the long four-minute format, in six well-trained sport students, warns that a fixed percentage of maximum heart rate produces a different stimulus in different people, and recommends individual lactate turn points instead. The same 85% is a different workout in two different bodies.

Intensity itself is not arbitrary. In a 2007 work-matched trial in 40 healthy, moderately trained men, 4x4 intervals at 90–95% of maximum heart rate raised VO2 max by 7.2%, significantly more than training at 85% or 70% with the same total oxygen cost. A pooled analysis of 18 interventions and 677 healthy, older and clinical participants found that high-volume intervals, more than 15 minutes of hard effort per session, the category the 4x4 belongs to, raised peak oxygen uptake by 3.3 mL/kg/min, against 2.4 for continuous training (P < 0.01) and 2.5 for low-volume or sprint intervals, a gap that was not significant (P = 0.09).

So the amount of hard work per session has an empirical advantage, and the exact edges of the protocol are a convention.

Causal direction

Partial. Evidence: moderate.

Both halves depend on one further question: whether exercise causes the benefit, or whether people with healthier arteries find it easier to exercise.

The Norwegian trial raises the question itself. Inside it, patients whose peak oxygen uptake rose more tended to show a larger fall in plaque fat (Spearman correlation −0.44, P = 0.009). The authors present that as a hypothesis. It is the kind of association randomisation exists to test, not to settle.

Mendelian randomisation, which uses genetic variation as a natural experiment, points toward cause for disease risk, with caveats. An exposome-wide analysis finds genetically proxied moderate-to-vigorous activity protective against several cardiometabolic diseases, a group that includes coronary artery disease, though its summary gives no coronary-specific estimate. A UK Biobank analysis finds genetically proxied fitness protective against heart failure (OR 0.79, 95% CI 0.63–0.99), with weaker genetic evidence for activity itself.

The result that summarises this page comes from a cohort with a 38-year gap. Fitness was measured in 8,986 male conscripts at about 18; the same men had their arteries imaged in middle age. Those in the top third for aerobic fitness had 22% lower odds of severe coronary narrowing, and those high in both fitness and strength 33% lower (OR 0.67, 95% CI 0.52–0.87), along a reversed J-shaped curve, with no consistent association with plaque in the carotid arteries.

Higher fitness goes with less coronary disease, and genetic studies point to a causal part for cardiometabolic disease more broadly. That does not show that it removes plaque already in the artery wall. One proposed mechanism did not appear where it was tested: 12 weeks of high-intensity intervals in 168 randomised patients with stable, well-treated coronary disease produced no sustained fall in any systemic inflammatory marker against standard care. No guideline or consensus statement among the sources endorses plaque regression from high-intensity training; the trial’s authors and the scoping review do.

The adjudication

Take the sentence apart and each piece gets a different answer.

The fitness half is right, with the exact protocol, at group level and under supervision. Its own trials add that individual response is hard to predict, that the advantage faded without supervision, and that a gentler programme did more for walking distance.

The plaque half is neither the fact the comment states nor the blank the question invites. One randomised trial compared supervised interval training with standard care using coronary imaging. Its main question, plaque fat, came back with no difference. Its volume measure favoured training by 1.4 percentage points of atheroma volume, in 60 patients on top of standard care, and we found no repeat against standard care. A second small trial found no difference between interval and continuous training, and imaging regression has not been shown to mean fewer events. That is a promising signal, not an established effect.

The counter-claim holds for calcium scores. Whether that calcium carries excess risk is not settled: most studies that tracked events found none, and the outcome studies that would decide it are still being called for.

On both sides of the thread, a real finding travels without the part that limits it.

Verdict: partial. VO2 max: supported, with the prescription used in the claim. Plaque regression: one randomised trial of interval training against standard care, a null main result and a small favourable volume result, not replicated; not established. The protocol’s numbers: a convention, around an intensity that has measured support. Causation: activity and fitness likely protective against cardiometabolic disease; reversal of existing plaque by training not established. The counter-claim: more calcium, yes; more events, not shown and not settled. And none of it was said by the four episodes under which the claim circulates.

98 published sources, from 99 records: one consensus statement appears in two journals. Of the 99, 15 were rated high strength, 53 moderate and 31 low. The opposing side holds 21 studies and 5 comments; the supporting side, 42 and 4. A random sample of 10 of the 98 DOIs resolved in Crossref: a sample, not a full check. Circulation from our collection round of 18/09/26, frozen on 24/09/26; source statements measured on 19/09/26. Sources are listed below by evidential hierarchy; the 2015 crossover study, which has no DOI, by its full title.