MedSysEvidence-graded reference

MedSys / Heart & cholesterol / Guidelines

Heart & cholesterol guide · Reference — not a chapter

Where the guidelines stand

The United States replaced its cholesterol guideline in March 2026 and the change was substantial. The United Kingdom did not. This page records what moved, what did not, and why the two cannot be read across.

Full guide Clinical pathway Eating for the heart

The short version

  • UK practice is unchanged. NICE NG238 (December 2023) is still the current standard. Its September 2025 review added links to technology appraisals; it did not change the recommendations.
  • The US guideline is new and it is a rewrite, not a refresh. Published 13 March 2026, it retires the 2018 document and is deliberately retitled from "blood cholesterol" to "dyslipidemia".
  • Europe sits between the two, with a 2025 focused update to the 2019 ESC/EAS guidelines.
  • Do not cross-apply the numbers. US targets are in mg/dL and rest on PREVENT; UK practice runs on QRISK3, mmol/L and non-HDL-C. The frameworks are not interchangeable.

Side by side

 United KingdomEuropeUnited States
Current document NICE NG238 ESC/EAS 2019, focused update 2025 ACC/AHA multisociety 2026
Date 14 Dec 2023
reviewed 2 Sep 2025		 7 Nov 2025 13 Mar 2026
Risk tool QRISK3, ages 25–84 SCORE2 / SCORE2-OP PREVENT-ASCVD, ages 30–79
replaces Pooled Cohort Equations
Units mmol/Lmmol/Lmg/dL
Primary treatment metric Non-HDL-C reduction LDL-C goals by risk category LDL-C and non-HDL-C goals restored
Lp(a) Selective, by risk Once in adult life Once in every adult's lifetime, plus cascade screening

SCORE2 and the European rows are included for orientation only; this site follows UK practice throughout, and the clinical pathway is written against NICE.

The targets, all converted to mmol/L

The US publishes in mg/dL, which makes side-by-side comparison awkward for a UK reader. Everything below is converted to mmol/L at 38.67 mg/dL per mmol/L. Where the 2026 US guideline states both units itself, the conversions here match its own figures exactly.

Situation UK — NICE NG238 Europe — ESC/EAS US — ACC/AHA 2026
Established CVD,
very high risk		 ≤ 2.0 LDL-C
or non-HDL-C ≤ 2.6		 < 1.4 LDL-C
plus ≥ 50% reduction from baseline		 < 1.4 LDL-C
and non-HDL-C < 2.2
Established CVD,
not very high risk		 ≤ 2.0 LDL-C
NICE does not sub-divide		 < 1.8 LDL-C < 1.8 LDL-C
Primary prevention,
high risk		 No absolute target
aim > 40% reduction in non-HDL-C		 < 1.8 LDL-C
plus ≥ 50% reduction		 < 1.8 LDL-C
HeFH, a risk factor, or subclinical atherosclerosis
Primary prevention,
moderate risk		 No absolute target < 2.6 LDL-C < 2.6 LDL-C
CAC 1–99 AU and below the 75th percentile
Primary prevention,
low risk		 No absolute target < 3.0 LDL-C —
CAC 100–299 AU
or ≥ 75th percentile		 CAC not used Not a target driver < 1.8 LDL-C
CAC ≥ 1000 AU CAC not used Not a target driver < 1.4 LDL-C
Lp(a) risk thresholds No stated threshold > 50 mg/dL ≥ 125 nmol/L ≈ 1.4× risk
≥ 250 nmol/L ≈ 2× risk
Severe hypercholesterolaemia FH: see NICE CG71 — ≥ 4.9 LDL-C
190 mg/dL; treat regardless of risk score

The one difference that actually matters

For a patient who has already had a heart attack or stroke and is at very high risk, the UK aims for LDL-C ≤ 2.0 mmol/L. Europe and the United States both aim for < 1.4 mmol/L. That is not a rounding difference — it is roughly a 30% lower target, and in a population of millions it is the gap that matters most.

The reasons are defensible rather than arbitrary. NICE weighs cost-effectiveness within the NHS, and the escalation needed to reach 1.4 mmol/L in most patients means ezetimibe, bempedoic acid, inclisiran or a PCSK9 inhibitor on top of high-intensity statin. NG238 also made the 2.0 figure a new and tighter recommendation in December 2023; before that there was no absolute secondary-prevention target in NICE guidance at all. The direction of travel is the same everywhere, the UK is simply further back along it.

Worth knowing in practice: the NHSE/AAC operational summary has carried the Joint British Societies figures of 1.8 and 2.5 mmol/L, which sit between NICE and the European position. If your local protocol quotes those, that is why.

United Kingdom — unchanged

NICE NG238, Cardiovascular disease: risk assessment and reduction, including lipid modification, was published on 14 December 2023, replacing CG181 from 2014. It was last reviewed on 2 September 2025, when links to relevant technology appraisal guidance were added — the recommendations themselves did not change. Anyone expecting a 2026 UK document should stop waiting for one.

The positions that matter in practice:

  • Risk assessment — QRISK3 for most adults aged 25 to 84 without established CVD.
  • Primary prevention — atorvastatin 20 mg where QRISK3 is 10% or above, after shared decision-making, aiming for a greater than 40% reduction in non-HDL cholesterol.
  • Secondary prevention — atorvastatin 80 mg unless a lower dose is indicated, targeting LDL-C ≤ 2.0 mmol/L or non-HDL-C ≤ 2.6 mmol/L. This was the substantive new recommendation in the December 2023 update.

Operationally, most trusts work from the NHS England / Accelerated Access Collaborative summary of national guidance for lipid management rather than from NG238 directly. Worth knowing: that summary historically carried the Joint British Societies targets of 1.8 and 2.5 mmol/L, which are lower than NICE's. If a local protocol quotes those figures, it may predate alignment with NG238 — check which version is in front of you.

United States — a rewrite

The 2026 ACC/AHA multisociety guideline was published simultaneously in JACC and Circulation on 13 March 2026, developed with nine further organisations. It retires the 2018 blood cholesterol guideline outright. The retitling is the clearest signal of intent: the subject is now atherogenic lipoproteins as a class — triglycerides, remnants and Lp(a) alongside LDL — rather than cholesterol alone.

What changed:

  • PREVENT-ASCVD replaces the Pooled Cohort Equations for primary prevention, giving both 10-year and 30-year estimates in adults aged 30 to 79. The 30-year horizon is the conceptually interesting part: it makes cumulative exposure explicit rather than implied.
  • Treatment goals return. The 2018 guideline moved away from numeric LDL-C targets toward percentage reduction; the 2026 document restores goals, graded by risk. Notably they are tied to imaging as well as risk score — any detectable coronary calcium supports an LDL-C goal below 100 mg/dL, falling to below 55 mg/dL where the calcium score reaches 1000 AU or above.
  • Lp(a) at least once in every adult's lifetime, with genetic cascade screening for relatives. This is the first time ACC/AHA has recommended universal Lp(a) screening, and it is the point of closest agreement with what this site has argued throughout.
  • apoB testing is endorsed for refining risk where the standard panel is ambiguous.
  • A life-course approach. Universal lipid screening for children aged 9 to 11 who have not previously been tested, and from age 2 where there is a family history of premature ASCVD, severe hypercholesterolaemia or known FH. In young adults, pharmacotherapy is considered at LDL-C of 160 mg/dL or above, or with a strong family history.

Europe — a focused update

The 2025 focused update to the 2019 ESC/EAS dyslipidaemia guidelines appeared in the European Heart Journal in November 2025. It is the document most likely to influence UK secondary care, since European practice shares both the unit system and much of the risk-assessment philosophy. The 2026 US guideline's summary materials include a direct comparison against it, which is the most efficient way to see where the two transatlantic positions diverge.

A correction to the focused update was published in the same journal in February 2026. Check it before citing specific figures.

Why you cannot simply read the US numbers across

  • Different units. mg/dL against mmol/L. LDL-C below 70 mg/dL is roughly 1.8 mmol/L; below 55 is roughly 1.4. Converting the number is easy; converting the justification is not.
  • Different risk engines. PREVENT and QRISK3 are built on different populations, include different variables, and are not calibrated to one another. A PREVENT percentage is not a QRISK3 percentage.
  • Different service context. NICE recommendations account for cost-effectiveness within the NHS. A guideline written without that constraint will reasonably reach different conclusions from the same evidence.

None of which makes the US document irrelevant to a UK reader. It reflects the same underlying evidence, and on Lp(a), apoB and cumulative exposure it has moved toward positions this site already took. But it is not the standard your GP is working to, and asking them to apply it would be asking them to depart from national guidance.

Sources

Reviewed August 2026. Guidelines move. If you are reading this a long way from that date, check NICE directly before relying on it. The sources above also appear on the site-wide reference list, alongside what each one does and does not support.