High cholesterol: what the guidelines say

High cholesterol, more precisely a high level of LDL cholesterol, the particle that carries cholesterol into artery walls, is one of the best-established, most directly modifiable contributors to atherosclerotic cardiovascular disease: heart attack, stroke, and related conditions caused by plaque buildup in arteries. In March 2026, the American College of Cardiology, American Heart Association, and nine co-sponsoring societies published a new dyslipidemia guideline that formally replaces the 2018 cholesterol guideline that had shaped care for years. This page describes what that current guideline says.

What “high” means, and why there is no single cutoff

Unlike blood pressure, cholesterol guidelines have not worked from a single diagnostic threshold for years, and the 2026 guideline continues that approach. LDL cholesterol is reported in mg/dL, and the guideline does define one clear high-severity cutoff, LDL-C of 190 mg/dL or above, which it calls severe hypercholesterolemia. Below that level, the guideline does not say a given LDL-C number is automatically “normal” or “abnormal” for treatment purposes. Instead, it asks a different question: given a person’s overall estimated risk of a future cardiovascular event, does lowering LDL-C, at whatever level it currently sits, produce enough benefit to be worth a medication decision. That is a genuinely different logic than a lab test with one reference range, and it is worth understanding before looking at any single LDL-C number on a lab report.

The major groups the guideline organizes care around

The guideline separates people into several groups, each with different guidance about how aggressively to lower LDL-C:

• People who already have clinical ASCVD (a prior heart attack, stroke, or related diagnosis): the guideline describes this as the highest-priority group, with distinct LDL-C treatment goals depending on how many high-risk features a person has.
• Severe hypercholesterolemia (LDL-C of 190 mg/dL or higher): the guideline describes this level as warranting attention on its own, independent of a calculated risk score, often because it reflects a genetic pattern of cholesterol handling.
• Adults with diabetes, generally ages 40 to 75, without an ASCVD diagnosis: the guideline describes this group as carrying elevated risk on the basis of diabetes itself, with the intensity of the suggested approach varying by how many additional risk factors are present.
• Primary prevention, risk-based: adults roughly 30 to 79 without diabetes, severe hypercholesterolemia, or known ASCVD, where the guideline calculates a 10-year risk estimate using a newer risk calculator, described next, to inform the conversation.
• Subclinical atherosclerosis, identified by a coronary artery calcium (CAC) scan: the guideline describes CAC results as a way to refine the decision when the risk-based estimate alone leaves genuine uncertainty.

The risk calculator and how it frames the treatment decision

The 2026 guideline drops the older Pooled Cohort Equations in favor of the newer PREVENT-ASCVD equations to estimate a person’s 10-year cardiovascular risk for the primary-prevention group. The guideline describes four risk bands from that calculation: low (under 3%), borderline (3% to under 5%), intermediate (5% to under 10%), and high (10% or above). In broad terms, the guideline describes a statin-based conversation as reasonable at the borderline level, and as recommended, following a discussion between clinician and patient, at intermediate and high levels. These percentage cutoffs are not directly comparable to the ones used under the older calculator: PREVENT tends to generate lower estimated risk for the same person than the tool it replaced, which is part of why the guideline’s numeric bands were recalibrated rather than left unchanged.

The guideline describes several factors that can shift the conversation even within a risk band, sometimes called risk-enhancing factors: a family history of premature cardiovascular disease, an elevated lipoprotein(a) level, an elevated hs-CRP level, and certain other markers. For people in the borderline or intermediate bands where the decision is genuinely close, the guideline describes a coronary artery calcium scan as a tool that can move the decision in either direction, including supporting a decision to hold off on medication when the scan shows no detectable calcium, and supporting a more intensive approach when it shows a substantial amount.

Medication classes the guideline describes

Consistent with prior guidelines, statins remain the guideline’s foundational drug class for lowering LDL-C, described at different intensities depending on the risk group and how much LDL-C reduction the situation calls for. For people who need additional LDL-C lowering beyond what a statin alone achieves, or who cannot tolerate a statin, the guideline describes several other drug classes: ezetimibe, a class of injectable PCSK9 inhibitors, and bempedoic acid, an oral non-statin option. None of these classes is dosed on this page; how much of any medication to use, and in what combination, is a clinical decision made between a person and their clinician based on their specific LDL-C level, risk profile, and tolerance.

The honest nuances

The single largest nuance here is that an entire guideline that had shaped cholesterol care since 2018 was replaced in 2026, and the change is more than cosmetic. Moving to PREVENT and recalibrating the risk bands means two people with identical lab values and history could, depending on exactly when they were evaluated, have received a different risk percentage and a different framing of the medication conversation under the old system versus the new one. That is not a sign either guideline was wrong; it reflects real, ongoing refinement of how cardiovascular risk is estimated. The guideline is also explicit that intermediate-risk decisions are meant to follow a clinician-patient discussion, not a rule applied automatically, and it introduces a formal decision process, sometimes described as calculate, personalize, reclassify, and reassess, specifically because individual circumstances (family history, lipoprotein(a), calcium scan results, and personal risk tolerance) can reasonably move two people with the same risk percentage toward different decisions.

Questions to discuss with your clinician

• What is my actual LDL-C number, and does it fall into the severe hypercholesterolemia range on its own, independent of a risk calculation?
• What is my 10-year risk estimate under the current PREVENT-ASCVD equations, and which risk band does that put me in?
• Do I have any of the risk-enhancing factors the guideline describes, such as a family history of early heart disease or an elevated lipoprotein(a) level, and were those factored into my risk estimate?
• Would a coronary artery calcium scan help clarify a decision that otherwise feels close, given my specific risk band?
• If a medication is being discussed, which class is being considered for my situation, and why that one over the alternatives the guideline describes?
• How does having diabetes, or a prior cardiovascular event, change which part of this guideline applies to me?

For related evidence on non-drug factors that affect cardiovascular risk, see dietary patterns and physical activity, or read the companion profile on high blood pressure.