The statin debate has ruined otherwise careful thinkers. The evidence base is one of the largest in cardiovascular medicine, and the risk/benefit is surprisingly well-characterized for a class of drugs that's been controversial in media coverage.
The headline meta-analysis
The Cholesterol Treatment Trialists (CTT) Collaboration has aggregated individual participant data across 27+ RCTs, including primary and secondary prevention trials (n > 186,000) ( Cholesterol Treatment Trialists Collaboration 2019, n=186854 ). Core finding: each 1 mmol/L (≈39 mg/dL) reduction in LDL-C via statin therapy produces:
- ~22% reduction in major vascular events.
- ~25% reduction in cardiovascular mortality in higher-risk groups.
- ~10% reduction in all-cause mortality in secondary prevention populations.
The relative risk reduction is proportional to the LDL reduction. Bigger cut = bigger benefit. This is mechanistically consistent with the cumulative-ApoB-exposure hypothesis: what matters is ApoB-years, not single-point ApoB.
For the complementary EPA intervention: REDUCE-IT (Bhatt 2019, n=8,179) demonstrated icosapent ethyl 4 g/day reduced major cardiovascular events 25% in statin-treated hypertriglyceridemic patients ( Bhatt et al. (REDUCE-IT) 2019, n=8179 ). Lipid-lowering beyond LDL does additional work in specific populations.
The primary prevention question
For someone with no prior cardiovascular event, the key question is not "do statins work" (they do) but "is the absolute benefit worth the side-effect and cost profile?"
Rough numbers for a 45-year-old with no CVD, no diabetes, moderate ApoB:
- 10-year CVD risk without statin: ~5%.
- With statin (25% reduction): ~3.75%.
- Absolute risk reduction: 1.25%.
- NNT to prevent one event over 10 years: ~80.
For a 60-year-old with ApoB 130 mg/dL + hypertension:
- 10-year CVD risk without statin: ~20%.
- With statin: ~15%.
- ARR: 5%.
- NNT: ~20.
Primary prevention benefit scales with baseline risk. That's why the decision should flow from a proper ASCVD risk calculation (pooled cohort equation, MESA + coronary calcium score where available), not a reflex prescription on any elevated LDL.
ApoB as the operational target
LDL-C measures cholesterol in LDL particles. ApoB counts the particles themselves. For the same LDL-C, one person can have twice as many ApoB particles as another, and it's the particle count that drives atherosclerosis.
| Phase | Dose | Notes |
|---|---|---|
| Primary prevention, low risk | ApoB < 80 mg/dL | Most adults; lifestyle first |
| Elevated risk (family history, metabolic syndrome) | ApoB < 70 | Lifestyle + consider statin earlier |
| Secondary prevention (post-MI/stroke) | ApoB < 60 | Statin + ezetimibe or PCSK9i as needed |
| Familial hypercholesterolemia | ApoB < 50 | Maximal therapy warranted; genetic testing if suspected |
Order ApoB alongside standard lipid panel annually; it's ~$20 add-on at most commercial labs.
Side effects
RCT data vs observational data diverge sharply on side-effect rates:
- RCT-attributable myalgia: ~5% (placebo rates ~2-4%; statin-specific attributable 1-3%).
- Observational/clinical myalgia reports: 10-20%.
- The gap is largely nocebo + nonspecific muscle soreness attributed to the statin.
Strategies for statin-intolerant patients:
- Low-dose statin (5-10 mg atorvastatin) + ezetimibe 10 mg: similar LDL reduction to 40 mg monotherapy, often tolerated when high-dose isn't.
- Rotate statin class: pitavastatin, pravastatin have different pharmacokinetics; ~50% of "statin intolerant" patients tolerate a different statin.
- CoQ10 supplementation for myalgia: modest trial support, low downside.
- Bempedoic acid: newer non-statin LDL-lowerer; works via liver-specific ATP-citrate lyase inhibition; no muscle side effects.
Real side effects to monitor:
- Transaminitis (ALT > 3× upper limit): clinically significant, rare.
- New-onset diabetes: small absolute increase, largest in high-dose potent statins. Benefit typically outweighs for high-risk patients.
- Rhabdomyolysis: very rare; much more common with cyclosporine or certain antibiotic combinations.
What statins don't do
- They don't extend lifespan meaningfully in young low-risk adults with normal ApoB. Mechanism of benefit requires atherosclerosis to suppress.
- They don't address triglycerides or HDL meaningfully; other interventions (omega-3, fibrates, lifestyle) target those.
- They don't reduce dementia risk in most RCTs (observational signals are confounded).
Counter-view
Aseem Malhotra and the "cholesterol skeptic" camp argue the meta-analysis data is overstated due to industry-funded trial design and selective endpoint reporting. The CTT data survives most of these critiques; the open peer review of the underlying individual-participant data is unusually transparent for pharmacotherapy research. Peter Attia takes an aggressive "treat ApoB early" stance even in low-risk patients; evidence-defensible for high-LDL-C patients, a stretch for average-LDL-C adults under 40.
