Beyond LDL-C: Understanding Cardiovascular Risk With PREVENT, ApoB, Lp(a), Non-HDL-C And Advanced Lipid Analysis GlobalRPH

GlobalRPh Clinical Technology

Beyond LDL-C: Understanding Cardiovascular Risk With PREVENT, ApoB, Lp(a), Non-HDL-C and Advanced Lipid Analysis

Introducing the GlobalRPh Cardiovascular Risk and Advanced Lipid Analyzer, a layered cardiovascular prevention tool that combines the standard lipid panel with advanced lipoprotein markers, derived calculations, kidney and metabolic context, coronary calcium, and the American Heart Association PREVENT equations.

Audience: Clinicians, pharmacists, trainees and medically sophisticated readersEvidence review: August 2026

Open the Cardiovascular Risk and Advanced Lipid Analyzer

Why a modern cardiovascular risk assessment has to look beyond one cholesterol number

For decades, lipid interpretation often centered on a single question: What is the LDL cholesterol? LDL-C remains one of the most important measurements in cardiovascular prevention, but it answers only part of the biological question. It estimates the amount of cholesterol carried inside LDL particles. It does not directly count the number of atherogenic particles, identify genetically elevated lipoprotein(a), characterize kidney-related risk, estimate long-term heart-failure risk, or determine whether coronary atherosclerosis is already present.

Modern prevention instead asks several related questions: How much atherogenic cholesterol is circulating? How many atherogenic particles are present? Is a largely inherited risk factor such as Lp(a) present? Are triglyceride-rich remnants contributing? How do blood pressure, diabetes, body size and kidney function alter absolute event risk? And, when uncertainty remains, is there evidence of existing coronary plaque?

Core principle: LDL-C, non-HDL-C, ApoB, Lp(a), PREVENT, metabolic and kidney markers, and coronary artery calcium do not compete to become the “best” number. They describe different parts of cardiovascular risk.

The GlobalRPh Cardiovascular Risk and Advanced Lipid Analyzer was designed around that layered approach. It can analyze partial or complete data, calculate derived lipid measures, identify missing high-yield information, compare cholesterol burden with particle burden, and calculate PREVENT risk when the required variables are available.

4Core measurements in the standard lipid panel

2High-yield additions: ApoB and Lp(a)

6PREVENT outputs: 10- and 30-year ASCVD, total CVD and HF

8+Derived lipid ratios and contextual calculations

Which laboratory tests should be considered?

Not every patient needs every cardiovascular biomarker. A more useful strategy is to start with the standard lipid phenotype, add high-yield markers that answer questions the standard panel cannot answer, then add metabolic, kidney or plaque information when it can change interpretation.

1. Standard lipid panel

Total cholesterol, LDL-C, HDL-C and triglycerides.

This is the foundation. It defines the basic lipid phenotype and permits calculation of non-HDL-C, remnant cholesterol and several supportive ratios.

2. High-yield additions

Lp(a) and ApoB.

Lp(a) can uncover largely inherited risk. ApoB estimates atherogenic particle number and is especially useful when LDL-C may not fully reflect particle burden.

3. Metabolic and kidney context

A1c or fasting glucose, eGFR, uACR and blood pressure.

These markers help characterize the cardiovascular-kidney-metabolic setting in which the lipid values should be interpreted.

4. Selective tests

hsCRP, ApoA-I, ApoB:ApoA-I and CAC.

These are not required for everyone. They are most useful when a specific residual-risk or treatment-decision question remains.

Have the laboratory report available?

Enter whatever data are available. The tool will analyze partial data and identify what additional information would materially improve the assessment.

Analyze the cardiovascular profile

How the standard lipid panel fits together

Total cholesterol

Total cholesterol is the cholesterol carried across the major circulating lipoprotein classes. It is useful for orientation, but treatment decisions are generally driven more directly by LDL-C, non-HDL-C, ApoB, absolute risk, clinical context and evidence of atherosclerosis.

LDL-C: cholesterol cargo, not particle count

LDL-C estimates the mass of cholesterol being transported within LDL particles. It remains a primary treatment target because cumulative exposure to atherogenic LDL particles is central to atherosclerosis. However, LDL-C does not literally count the number of LDL or other ApoB-containing particles.

Useful analogy: Think of ApoB-containing lipoproteins as vehicles moving cholesterol through the bloodstream. LDL-C describes part of the cholesterol cargo. ApoB is closer to counting the number of atherogenic vehicles.

HDL-C

HDL-C measures cholesterol carried in high-density lipoproteins. Low HDL-C often accompanies insulin resistance, hypertriglyceridemia and metabolic syndrome. High HDL-C, however, should not be treated as a shield that cancels high LDL-C, high ApoB, high Lp(a), smoking, diabetes, CKD or established plaque. HDL-C is useful for risk assessment and pattern recognition, but it is not a treatment target simply because raising the laboratory value appears desirable.

Triglycerides

Triglycerides circulate mainly within triglyceride-rich lipoproteins. Persistent elevation can occur with insulin resistance, diabetes, obesity, alcohol exposure, medications, kidney disease, hypothyroidism and other conditions. Elevated triglycerides can also indicate increased remnant-lipoprotein burden. At very high concentrations, pancreatitis prevention becomes an additional immediate clinical priority.

Non-HDL cholesterol: one of the most useful calculations already hidden in a routine lipid panel

FormulaNon-HDL-C = Total cholesterol − HDL-C

Non-HDL-C represents cholesterol carried in essentially all lipoprotein classes other than HDL. That includes LDL, IDL, VLDL and VLDL remnants, chylomicron remnants when present, and Lp(a). In practical terms, it captures cholesterol across the broader family of potentially atherogenic ApoB-containing particles.

This is particularly useful when triglycerides are elevated. In that setting, a meaningful portion of atherogenic cholesterol may be traveling in triglyceride-rich particles and their remnants rather than in conventional LDL alone.

Worked exampleTotal cholesterol 205 − HDL-C 55 = non-HDL-C 150 mg/dL

Non-HDL-C versus ApoB

The measurements are complementary. Non-HDL-C measures cholesterol mass. ApoB estimates particle number. Two people can therefore have the same non-HDL-C yet have different numbers of circulating atherogenic particles.

The 2026 ACC/AHA dyslipidemia guideline restored LDL-C and non-HDL-C treatment goals, reinforcing the clinical importance of evaluating cholesterol beyond LDL-C alone.^[1]

Do not confuse a descriptive range with a treatment goal. Appropriate LDL-C and non-HDL-C targets depend on the prevention category. The calculator uses risk-aligned goal context rather than assuming that one “normal” cutoff applies to everyone.

ApoB: estimating the number of atherogenic particles

Apolipoprotein B is the main structural protein on the major atherogenic lipoprotein particles. Each circulating LDL, IDL, VLDL/remnant and Lp(a) particle carries one ApoB molecule. ApoB concentration can therefore be used as an estimate of the number of particles capable of interacting with the arterial wall.

LDL-C asks

How much cholesterol is being carried inside LDL particles?

ApoB asks

Approximately how many atherogenic ApoB-containing particles are circulating?

Why LDL-C and ApoB can disagree

Particles do not all carry the same amount of cholesterol. Two patients with LDL-C of 100 mg/dL can have different ApoB concentrations. One may have fewer, cholesterol-richer particles. Another may have more numerous, relatively cholesterol-depleted particles. The second patient can therefore have greater particle burden despite the same LDL-C.

Discordance becomes particularly relevant with elevated triglycerides, diabetes, obesity, insulin resistance, CKD and very low LDL-C achieved during therapy. The 2026 guideline gives ApoB a larger role in assessing residual atherogenic risk in these settings.^[1]

Why the calculator includes ApoB/LDL-C and LDL-C/ApoB: These reciprocal ratios are not treatment targets. They are clues to whether particle number appears disproportionately high or low relative to cholesterol content.

Lipoprotein(a): inherited risk that the standard lipid panel cannot reveal

Lp(a) is an LDL-like particle with an additional apolipoprotein(a) attached. Its concentration is determined predominantly by genetics, which is why it can remain elevated even when diet, exercise and other lifestyle measures are favorable.

The 2026 dyslipidemia guideline recommends measuring Lp(a) at least once in adulthood. Levels at or above approximately 125 nmol/L or 50 mg/dL are considered risk enhancing, with higher levels carrying progressively greater long-term risk.^[1]

Do not convert Lp(a) with one fixed factor. Laboratories may report nmol/L or mg/dL. Because apolipoprotein(a) isoform size varies, there is no single exact conversion that applies to every patient. Use the unit reported by the laboratory.

Elevated Lp(a) does not automatically dictate one medication decision. Its practical importance is that it changes the overall risk conversation and increases the importance of controlling modifiable risk factors, particularly LDL-related exposure.

PREVENT: moving from isolated laboratory values to absolute cardiovascular risk

The American Heart Association PREVENT equations were developed from contemporary data involving more than 6.5 million diverse US adults. They estimate cardiovascular risk using cardiovascular, kidney and metabolic factors and do not use race as a biological predictor.^[2,3]

The GlobalRPh calculator uses the published base PREVENT equations to estimate:

PREVENT-ASCVD

Atherosclerotic events such as myocardial infarction and stroke.

PREVENT-total CVD

A broader cardiovascular outcome that includes ASCVD and heart failure.

PREVENT-HF

Incident heart failure risk, which is strongly influenced by CKM factors.

Ten-year estimates are used for eligible adults ages 30 to 79. The calculator displays 30-year estimates for adults ages 30 to 59, the validated long-term range highlighted by AHA resources.^[2,3]

Inputs required by the PHP calculator

Domain	Input	Why it matters
Demographic	Age and sex used by the published equations	PREVENT uses sex-specific coefficient sets and age interactions.
Lipids	Total cholesterol and HDL-C	The PHP code derives non-HDL-C internally and scales both non-HDL-C and HDL-C.
Blood pressure	Systolic BP and BP-treatment status	SBP is modeled with spline-like terms plus a treatment interaction.
Metabolic	BMI and diabetes status	These are especially important to total CVD and HF risk.
Kidney	eGFR	Kidney function is a core feature distinguishing PREVENT from older models.
Behavior/treatment	Current smoking and statin use	Both are modeled directly; statin use also interacts with non-HDL-C.

Important distinction: A1c, fasting glucose and uACR are clinically important fields in the GlobalRPh analyzer, but the v1.9 PHP calculator uses the published base PREVENT equations. Those additional values are interpreted separately rather than silently inserted into the base PREVENT percentage.

The actual PREVENT equation used by the PHP tool

The PHP calculator first transforms the raw inputs into the centered and scaled terms used in the published equations. The coefficient set then changes according to sex, prediction horizon and outcome. In other words, there is a distinct coefficient set for female and male 10-year ASCVD, total CVD and HF, and another set for the corresponding 30-year outcomes.

Step 1: lipid conversion and centered variables non-HDL-C (mmol/L) = (Total cholesterol − HDL-C) × 0.02586 HDL-C (mmol/L) = HDL-C (mg/dL) × 0.02586 A = (Age − 55) / 10 A² = A × A N = non-HDL-C (mmol/L) − 3.5 H = [HDL-C (mmol/L) − 1.3] / 0.3

Step 2: piecewise SBP, BMI and eGFR terms S₁ = [min(SBP,110) − 110] / 20 S₂ = [max(SBP,110) − 130] / 20 B₁ = [min(BMI,30) − 25] / 5 B₂ = [max(BMI,30) − 30] / 5 E₁ = [min(eGFR,60) − 60] / −15 E₂ = [max(eGFR,60) − 90] / −15

Step 3: linear predictor LP = β₀ + βageA + βage²A² + βNHN + βHDLH + βSBP1S₁ + βSBP2S₂ + βDM(DM) + βSMK(SMK) + βBMI1B₁ + βBMI2B₂ + βeGFR1E₁ + βeGFR2E₂ + βBPTx(BPTx) + βstatin(STATIN) + βTxSBP(BPTx×S₂) + βstatinNH(STATIN×N) + βageNH(A×N) + βageHDL(A×H) + βageSBP(A×S₂) + βageDM(A×DM) + βageSMK(A×SMK) + βageBMI(A×B₂) + βageeGFR(A×E₁)

Step 4: convert the linear predictor to predicted riskPredicted risk (%) = 100 × e^LP / (1 + e^LP)

DM, SMK, BPTx and STATIN are binary 0/1 variables. Some terms have a coefficient of zero in certain outcome models. For example, the heart-failure equations do not use the lipid terms in the same way the ASCVD equations do. The generic PHP function nevertheless applies one common model structure and the coefficient table determines which terms contribute.

PREVENT-ASCVD risk categories used by the calculator

10-year PREVENT-ASCVD	Calculator category	2026 context
<3%	Low	Emphasize health behaviors and reassessment; long-term risk can still matter in younger adults.
3% to <5%	Borderline	Personalize with risk enhancers and consider CAC selectively when a treatment decision remains uncertain.
5% to <10%	Intermediate	Risk-based lipid-lowering treatment is generally supported after clinician-patient review.
≥10%	High	Supports more intensive LDL-lowering context in the 2026 primary-prevention pathway.

The calculator applies PREVENT only when the required data are complete and when known ASCVD or severe subclinical cardiovascular disease has not already shifted the patient into a different prevention framework. The PHP implementation also warns that the 2026 risk-based lipid treatment pathway primarily applies PREVENT-ASCVD when LDL-C is 70 to 189 mg/dL.

Calculate, Personalize, Reclassify

The 2026 dyslipidemia guideline summarizes modern primary-prevention use of PREVENT with a useful CPR framework: Calculate PREVENT risk, Personalize the estimate with risk enhancers that are not fully represented in the base equations, and, when appropriate, Reclassify with selective coronary artery calcium testing.^[1]

See the PREVENT outputs together with the laboratory interpretation

The GlobalRPh analyzer presents PREVENT-ASCVD, total CVD and HF alongside the lipid and CKM findings rather than reducing the report to one percentage.

Calculate PREVENT risk

Technical appendix: exact coefficient sets in the v1.9 PHP calculator

The collapsible tables below reproduce the coefficient sets currently stored in the PHP calculator for each prediction horizon, outcome and sex. They are included for transparency and technical review. Most readers do not need these raw coefficients to interpret a PREVENT result.

10-year ASCVD — Female

Model term	Coefficient
Intercept	`-3.819975`
Age	`0.719883`
Age²	`0`
Non-HDL-C	`0.1176967`
HDL-C	`-0.151185`
SBP lower spline	`-0.0835358`
SBP upper spline	`0.3592852`
Diabetes	`0.8348585`
Current smoking	`0.4831078`
BMI lower spline	`0`
BMI upper spline	`0`
eGFR lower spline	`0.4864619`
eGFR upper spline	`0.0397779`
BP treatment	`0.2265309`
Statin use	`-0.0592374`
BP treatment × SBP	`-0.0395762`
Statin × non-HDL-C	`0.0844423`
Age × non-HDL-C	`-0.0567839`
Age × HDL-C	`0.0325692`
Age × SBP	`-0.1035985`
Age × diabetes	`-0.2417542`
Age × smoking	`-0.0791142`
Age × BMI	`0`
Age × eGFR	`-0.1671492`

10-year ASCVD — Male

Model term	Coefficient
Intercept	`-3.500655`
Age	`0.7099847`
Age²	`0`
Non-HDL-C	`0.1658663`
HDL-C	`-0.1144285`
SBP lower spline	`-0.2837212`
SBP upper spline	`0.3239977`
Diabetes	`0.7189597`
Current smoking	`0.3956973`
BMI lower spline	`0`
BMI upper spline	`0`
eGFR lower spline	`0.3690075`
eGFR upper spline	`0.0203619`
BP treatment	`0.2036522`
Statin use	`-0.0865581`
BP treatment × SBP	`-0.0322916`
Statin × non-HDL-C	`0.114563`
Age × non-HDL-C	`-0.0300005`
Age × HDL-C	`0.0232747`
Age × SBP	`-0.0927024`
Age × diabetes	`-0.2018525`
Age × smoking	`-0.0970527`
Age × BMI	`0`
Age × eGFR	`-0.1217081`

10-year total CVD — Female

Model term	Coefficient
Intercept	`-3.307728`
Age	`0.7939329`
Age²	`0`
Non-HDL-C	`0.0305239`
HDL-C	`-0.1606857`
SBP lower spline	`-0.2394003`
SBP upper spline	`0.3600781`
Diabetes	`0.8667604`
Current smoking	`0.5360739`
BMI lower spline	`0`
BMI upper spline	`0`
eGFR lower spline	`0.6045917`
eGFR upper spline	`0.0433769`
BP treatment	`0.3151672`
Statin use	`-0.1477655`
BP treatment × SBP	`-0.0663612`
Statin × non-HDL-C	`0.1197879`
Age × non-HDL-C	`-0.0819715`
Age × HDL-C	`0.0306769`
Age × SBP	`-0.0946348`
Age × diabetes	`-0.27057`
Age × smoking	`-0.078715`
Age × BMI	`0`
Age × eGFR	`-0.1637806`

10-year total CVD — Male

Model term	Coefficient
Intercept	`-3.031168`
Age	`0.7688528`
Age²	`0`
Non-HDL-C	`0.0736174`
HDL-C	`-0.0954431`
SBP lower spline	`-0.4347345`
SBP upper spline	`0.3362658`
Diabetes	`0.7692857`
Current smoking	`0.4386871`
BMI lower spline	`0`
BMI upper spline	`0`
eGFR lower spline	`0.5378979`
eGFR upper spline	`0.0164827`
BP treatment	`0.288879`
Statin use	`-0.1337349`
BP treatment × SBP	`-0.0475924`
Statin × non-HDL-C	`0.150273`
Age × non-HDL-C	`-0.0517874`
Age × HDL-C	`0.0191169`
Age × SBP	`-0.1049477`
Age × diabetes	`-0.2251948`
Age × smoking	`-0.0895067`
Age × BMI	`0`
Age × eGFR	`-0.1543702`

10-year HF — Female

Model term	Coefficient
Intercept	`-4.310409`
Age	`0.8998235`
Age²	`0`
Non-HDL-C	`0`
HDL-C	`0`
SBP lower spline	`-0.4559771`
SBP upper spline	`0.3576505`
Diabetes	`1.038346`
Current smoking	`0.583916`
BMI lower spline	`-0.0072294`
BMI upper spline	`0.2997706`
eGFR lower spline	`0.7451638`
eGFR upper spline	`0.0557087`
BP treatment	`0.3534442`
Statin use	`0`
BP treatment × SBP	`-0.0981511`
Statin × non-HDL-C	`0`
Age × non-HDL-C	`0`
Age × HDL-C	`0`
Age × SBP	`-0.0946663`
Age × diabetes	`-0.3581041`
Age × smoking	`-0.1159453`
Age × BMI	`-0.003878`
Age × eGFR	`-0.1884289`

10-year HF — Male

Model term	Coefficient
Intercept	`-3.946391`
Age	`0.8972642`
Age²	`0`
Non-HDL-C	`0`
HDL-C	`0`
SBP lower spline	`-0.6811466`
SBP upper spline	`0.3634461`
Diabetes	`0.923776`
Current smoking	`0.5023736`
BMI lower spline	`-0.0485841`
BMI upper spline	`0.3726929`
eGFR lower spline	`0.6926917`
eGFR upper spline	`0.0251827`
BP treatment	`0.2980922`
Statin use	`0`
BP treatment × SBP	`-0.0497731`
Statin × non-HDL-C	`0`
Age × non-HDL-C	`0`
Age × HDL-C	`0`
Age × SBP	`-0.1289201`
Age × diabetes	`-0.3040924`
Age × smoking	`-0.1401688`
Age × BMI	`0.0068126`
Age × eGFR	`-0.1797778`

30-year ASCVD — Female

Model term	Coefficient
Intercept	`-1.974074`
Age	`0.4669202`
Age²	`-0.0893118`
Non-HDL-C	`0.1256901`
HDL-C	`-0.1542255`
SBP lower spline	`-0.0018093`
SBP upper spline	`0.322949`
Diabetes	`0.6296707`
Current smoking	`0.268292`
BMI lower spline	`0`
BMI upper spline	`0`
eGFR lower spline	`0.100106`
eGFR upper spline	`0.0499663`
BP treatment	`0.1875292`
Statin use	`0.0152476`
BP treatment × SBP	`-0.0276123`
Statin × non-HDL-C	`0.0736147`
Age × non-HDL-C	`-0.0521962`
Age × HDL-C	`0.0316918`
Age × SBP	`-0.1046101`
Age × diabetes	`-0.2727793`
Age × smoking	`-0.1530907`
Age × BMI	`0`
Age × eGFR	`-0.1299149`

30-year ASCVD — Male

Model term	Coefficient
Intercept	`-1.736444`
Age	`0.3994099`
Age²	`-0.0937484`
Non-HDL-C	`0.1744643`
HDL-C	`-0.120203`
SBP lower spline	`-0.0665117`
SBP upper spline	`0.2753037`
Diabetes	`0.4790257`
Current smoking	`0.1782635`
BMI lower spline	`0`
BMI upper spline	`0`
eGFR lower spline	`-0.0218789`
eGFR upper spline	`0.0602553`
BP treatment	`0.1421182`
Statin use	`0.0135996`
BP treatment × SBP	`-0.0218265`
Statin × non-HDL-C	`0.1013148`
Age × non-HDL-C	`-0.0312619`
Age × HDL-C	`0.020673`
Age × SBP	`-0.0920935`
Age × diabetes	`-0.2159947`
Age × smoking	`-0.1548811`
Age × BMI	`0`
Age × eGFR	`-0.0712547`

30-year total CVD — Female

Model term	Coefficient
Intercept	`-1.318827`
Age	`0.5503079`
Age²	`-0.0928369`
Non-HDL-C	`0.0409794`
HDL-C	`-0.1663306`
SBP lower spline	`-0.1628654`
SBP upper spline	`0.3299505`
Diabetes	`0.6793894`
Current smoking	`0.3196112`
BMI lower spline	`0`
BMI upper spline	`0`
eGFR lower spline	`0.1857101`
eGFR upper spline	`0.0553528`
BP treatment	`0.2894`
Statin use	`-0.075688`
BP treatment × SBP	`-0.056367`
Statin × non-HDL-C	`0.1071019`
Age × non-HDL-C	`-0.0751438`
Age × HDL-C	`0.0301786`
Age × SBP	`-0.0998776`
Age × diabetes	`-0.3206166`
Age × smoking	`-0.1607862`
Age × BMI	`0`
Age × eGFR	`-0.1450788`

30-year total CVD — Male

Model term	Coefficient
Intercept	`-1.148204`
Age	`0.4627309`
Age²	`-0.0984281`
Non-HDL-C	`0.0836088`
HDL-C	`-0.1029824`
SBP lower spline	`-0.2140352`
SBP upper spline	`0.2904325`
Diabetes	`0.5331276`
Current smoking	`0.2141914`
BMI lower spline	`0`
BMI upper spline	`0`
eGFR lower spline	`0.1155556`
eGFR upper spline	`0.0603775`
BP treatment	`0.232714`
Statin use	`-0.0272112`
BP treatment × SBP	`-0.0384488`
Statin × non-HDL-C	`0.134192`
Age × non-HDL-C	`-0.0511759`
Age × HDL-C	`0.0165865`
Age × SBP	`-0.1101437`
Age × diabetes	`-0.2585943`
Age × smoking	`-0.1566406`
Age × BMI	`0`
Age × eGFR	`-0.1166776`

30-year HF — Female

Model term	Coefficient
Intercept	`-2.205379`
Age	`0.6254374`
Age²	`-0.0983038`
Non-HDL-C	`0`
HDL-C	`0`
SBP lower spline	`-0.3919241`
SBP upper spline	`0.3142295`
Diabetes	`0.8330787`
Current smoking	`0.3438651`
BMI lower spline	`0.0594874`
BMI upper spline	`0.2525536`
eGFR lower spline	`0.2981642`
eGFR upper spline	`0.0667159`
BP treatment	`0.333921`
Statin use	`0`
BP treatment × SBP	`-0.0893177`
Statin × non-HDL-C	`0`
Age × non-HDL-C	`0`
Age × HDL-C	`0`
Age × SBP	`-0.0974299`
Age × diabetes	`-0.404855`
Age × smoking	`-0.1982991`
Age × BMI	`-0.0035619`
Age × eGFR	`-0.1564215`

30-year HF — Male

Model term	Coefficient
Intercept	`-1.95751`
Age	`0.5681541`
Age²	`-0.1048388`
Non-HDL-C	`0`
HDL-C	`0`
SBP lower spline	`-0.4761564`
SBP upper spline	`0.30324`
Diabetes	`0.6840338`
Current smoking	`0.2656273`
BMI lower spline	`0.0833107`
BMI upper spline	`0.26999`
eGFR lower spline	`0.2541805`
eGFR upper spline	`0.0638923`
BP treatment	`0.2583631`
Statin use	`0`
BP treatment × SBP	`-0.0391938`
Statin × non-HDL-C	`0`
Age × non-HDL-C	`0`
Age × HDL-C	`0`
Age × SBP	`-0.1269124`
Age × diabetes	`-0.3273572`
Age × smoking	`-0.2043019`
Age × BMI	`-0.0182831`
Age × eGFR	`-0.1342618`

What the cardiovascular ratios mean, and what they do not mean

The calculator derives several ratios because they can reveal patterns that are less obvious when each laboratory value is viewed separately. They should be treated as supportive pattern clues, not as competing risk scores and not as substitutes for LDL-C, non-HDL-C, ApoB, Lp(a), PREVENT or documented atherosclerosis.

Calculation	Formula	Primary use	Important limitation
Castelli I	Total cholesterol / HDL-C	Broad cholesterol-to-HDL balance	A favorable ratio can hide a high absolute LDL-C.
Castelli II	LDL-C / HDL-C	LDL-to-HDL pattern	Not a treatment target; absolute LDL-C remains more actionable.
TG / HDL-C	Triglycerides / HDL-C	Cardiometabolic and triglyceride-rich pattern clue	Not a stand-alone diagnostic test for insulin resistance.
AIP	log10[TG mmol/L / HDL-C mmol/L]	Exploratory atherogenic dyslipidemia marker	Not a routine US guideline treatment target.
Non-HDL-C / HDL-C	Non-HDL-C / HDL-C	Simple balance of atherogenic cholesterol to HDL-C	No universal treatment cutoff.
ApoB / ApoA-I	ApoB / ApoA-I	Balance of atherogenic particle burden and HDL-associated protein biology	ApoB alone is generally more actionable.
ApoB / LDL-C	ApoB / LDL-C	Clue to particle-number/cholesterol discordance	No universal treatment cutoff.
LDL-C / ApoB	LDL-C / ApoB	Rough clue to cholesterol carried per ApoB particle	Not a direct particle-size measurement.

Castelli Risk Index I: total cholesterol / HDL-C

The calculator describes values below 4 as generally favorable, 4 to below 5 as a middle range and 5 or greater as less favorable. These are descriptive bands, not guideline treatment thresholds.

Castelli Risk Index II: LDL-C / HDL-C

The tool uses below 2.5 as generally favorable, 2.5 to below 3.5 as intermediate and 3.5 or greater as less favorable. Again, the absolute LDL-C and overall risk take precedence.

Triglyceride / HDL-C ratio

A higher TG/HDL-C ratio often accompanies insulin resistance, hypertriglyceridemia, lower HDL-C and smaller LDL-particle patterns. The calculator uses below 2 as generally favorable, 2 to below 3 as borderline and 3 or greater as less favorable. Cohort evidence supports an association with cardiovascular outcomes, but interpretation varies by fasting status, sex, ancestry, diabetes and medications.^[7]

Atherogenic Index of Plasma

AIP formulaAIP = log10[TG (mmol/L) / HDL-C (mmol/L)]TG mmol/L = TG mg/dL × 0.01129 | HDL-C mmol/L = HDL-C mg/dL × 0.02586

The calculator uses AIP below 0.11 as a lower range, 0.11 to 0.21 as intermediate and above 0.21 as higher. Research links AIP with atherogenic dyslipidemia, smaller LDL particles and coronary disease burden, but it remains an exploratory or supportive marker rather than a current ACC/AHA treatment target.^[8]

ApoB:ApoA-I ratio

ApoA-I is the principal structural protein of HDL particles. The ApoB:ApoA-I ratio therefore compares atherogenic particle burden with HDL-associated ApoA-I biology. INTERHEART found the ratio strongly associated with myocardial infarction across diverse populations.^[6] The ratio is supplementary; a favorable value does not erase elevated LDL-C, ApoB, Lp(a) or established plaque.

Most important safeguard: High HDL-C can mathematically improve several ratios. That does not mean high HDL-C neutralizes a markedly elevated LDL-C, ApoB, Lp(a), smoking exposure, diabetes or coronary calcium.

Calculated remnant cholesterol

FormulaRemnant cholesterol = Total cholesterol − LDL-C − HDL-C

This value approximates cholesterol carried in triglyceride-rich remnant particles after LDL-C and HDL-C are subtracted from total cholesterol. It can be helpful in mixed or triglyceride-rich dyslipidemia.

It is not equivalent to a direct laboratory measurement of remnant lipoproteins. It also inherits error from the LDL-C value. If LDL-C is estimated, that estimation error is carried into calculated remnant cholesterol.

Worked exampleTotal cholesterol 200 − LDL-C 120 − HDL-C 50 = remnant cholesterol 30 mg/dL

When LDL-C is missing: the Sampson-NIH equation used by the tool

If LDL-C is not entered but total cholesterol, HDL-C and triglycerides are available, the PHP calculator estimates LDL-C with the Sampson-NIH equation through triglycerides of 800 mg/dL. The method was developed to improve LDL-C estimation, including in hypertriglyceridemic settings where the traditional Friedewald approach becomes less reliable.^[5]

Sampson-NIH equation implemented in the PHP calculatorLDL-C = TC/0.948 − HDL-C/0.971 − [TG/8.56 + (TG × non-HDL-C)/2140 − TG²/16100] − 9.44

The calculator does not apply this equation when triglycerides exceed 800 mg/dL. When an appropriate laboratory-reported LDL-C is available, the reported value is generally preferable to generating a replacement estimate.

BMI, glycemia and kidney function: why cardiovascular risk is now a CKM problem

The 2026 AHA/ACC/ADA/ASN cardiovascular-kidney-metabolic guideline formalizes the idea that obesity, diabetes, kidney disease and cardiovascular disease are biologically interconnected rather than separate silos. PREVENT plays a central role in risk assessment across CKM stages 0 through 3.^[4]

BMI

FormulaBMI = weight in kilograms / height in meters²Pounds × 0.453592 = kilograms

The calculator can accept feet/inches and pounds or metric units and automatically derives BMI. BMI is a screening measure. It does not directly measure visceral fat, body-fat percentage, skeletal-muscle mass, fat distribution, cardiorespiratory fitness or insulin sensitivity.

eGFR and uACR answer different kidney questions

eGFR primarily characterizes filtration function. Urine albumin-creatinine ratio characterizes albuminuria and kidney damage. A patient can have relatively preserved eGFR yet have clinically important albuminuria, so the two measurements are complementary.

A1c and fasting glucose

Diabetes status is a required binary PREVENT input, but A1c and fasting glucose provide additional information about current glycemic burden. The GlobalRPh tool reviews them separately so that prediabetes-range abnormalities or worsening glycemia are not lost simply because the PREVENT diabetes field is yes/no.

Coronary artery calcium and hsCRP answer different questions

CAC: moving from predicted risk toward evidence of plaque

A coronary artery calcium score is obtained by noncontrast CT and quantifies calcified coronary plaque. PREVENT estimates the probability of future events. CAC can demonstrate that calcified coronary atherosclerosis is already present.

The 2026 dyslipidemia guideline expands selective use of CAC when risk-based treatment decisions remain uncertain. A score of zero can be reassuring in an appropriate primary-prevention setting, but it does not exclude noncalcified plaque and does not nullify every other risk factor.^[1]

hsCRP

High-sensitivity C-reactive protein can provide inflammation-related risk context when persistently elevated in a clinically stable patient. It is nonspecific. Infection, inflammatory disease, trauma and many other conditions can markedly elevate CRP, so values around or above 10 mg/L generally require reassessment when the patient is stable before they are interpreted as cardiovascular-risk information.

Common mistakes when interpreting advanced cardiovascular results

“My HDL is high, so my LDL does not matter.”

Incorrect. High HDL-C does not erase the risk associated with high atherogenic-lipoprotein exposure.

“Normal LDL-C means ApoB must be normal.”

Not necessarily. Cholesterol mass and particle number can be discordant.

“My ratio is excellent, therefore my risk is low.”

Not necessarily. Ratios can look favorable because of a high denominator while absolute LDL-C or ApoB remains important.

“My Lp(a) must be fine because my lipid panel is fine.”

No. Lp(a) must be measured separately.

“A low PREVENT score proves I have no plaque.”

No. PREVENT predicts events. It does not image the coronary arteries.

“CAC = 0 means no atherosclerosis exists.”

Not necessarily. CAC detects calcified plaque and does not exclude noncalcified plaque.

“Remnant cholesterol was directly measured.”

Not when it is derived by subtraction. The value inherits uncertainty from LDL-C.

“A normal BMI means normal metabolic health.”

No. BMI does not directly measure visceral adiposity, fitness or insulin resistance.

Why the calculator analyzes missing data instead of requiring a perfect dataset

Real-world laboratory reports are often incomplete. A useful cardiovascular tool should therefore distinguish between what can be interpreted now and what remains unknown.

If only LDL-C is available, the analyzer can interpret LDL-C but cannot characterize non-HDL-C, HDL-related ratios, triglyceride-rich patterns, ApoB particle burden, Lp(a)-related inherited risk, PREVENT risk, CKM context or coronary plaque. If the standard lipid panel is complete but ApoB or Lp(a) is missing, the tool identifies those limitations separately. If PREVENT inputs are incomplete, it lists the missing variables rather than producing a false score.

This is intentional: “Unable to calculate” is less useful than explaining exactly which missing data limit the assessment and why those data might matter.

A practical hierarchy for interpreting the complete profile

Establish the standard lipid phenotype: total cholesterol, LDL-C, HDL-C and triglycerides.
Calculate broader atherogenic cholesterol: non-HDL-C.
Assess particle number when useful: ApoB.
Look for inherited risk: Lp(a), generally at least once in adulthood.
Review triglyceride-rich and cardiometabolic patterns: triglycerides, remnant cholesterol, TG/HDL-C and AIP.
Add CKM context: BMI, glycemia, eGFR, uACR and blood pressure.
Calculate absolute risk when appropriate: PREVENT-ASCVD, total CVD and HF.
Personalize the estimate: family history, inflammatory conditions, Lp(a), hsCRP and other risk enhancers.
Reclassify selectively: CAC when a treatment decision remains uncertain.
Change the framework when disease is already established: known ASCVD belongs to secondary prevention, not ordinary primary-prevention risk scoring.

Key takeaways

LDL-C remains fundamental, but it measures cholesterol content rather than total atherogenic particle number.
Non-HDL-C is available from a routine panel and captures cholesterol across a broader range of atherogenic particles.
ApoB estimates particle concentration and is especially useful when LDL-C may be discordant with particle burden.
Lp(a) identifies largely inherited risk that cannot be reliably inferred from an ordinary lipid panel.
PREVENT integrates cardiovascular, kidney and metabolic information and separates ASCVD, total CVD and heart-failure risk.
Thirty-year risk matters in younger adults because low short-term risk can coexist with substantial cumulative exposure.
Ratios help describe the pattern, but they should not override validated treatment targets or absolute risk.
CAC provides different information because it can demonstrate calcified coronary plaque rather than merely predict risk.

Put the pieces together in one report

The GlobalRPh Cardiovascular Risk and Advanced Lipid Analyzer combines the available lipid profile, advanced markers, derived calculations, PREVENT outputs, CKM context, risk enhancers and CAC information into one structured analysis.

Open the GlobalRPh Cardiovascular Risk and Advanced Lipid Analyzer

References

Blumenthal RS, Morris PB, Gaudino M, et al. 2026 ACC/AHA/AACVPR/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Dyslipidemia. Circulation and Journal of the American College of Cardiology. Published March 13, 2026. American Heart Association guideline hub.
Khan SS, Matsushita K, Sang Y, et al. Development and Validation of the American Heart Association’s PREVENT Equations. Circulation. 2024;149(6):430-449. doi:10.1161/CIRCULATIONAHA.123.067626. PubMed.
American Heart Association. Predicting Risk of cardiovascular disease EVENTs (PREVENT) Calculator and clinical resources. Professional Heart Daily.
Ndumele CE, Rodriguez F, Dixon DL, et al. 2026 AHA/ACC/ADA/ASN Guideline for the Prevention, Detection, Evaluation, and Management of Cardiovascular-Kidney-Metabolic Syndrome. Circulation. Published June 9, 2026. doi:10.1161/CIR.0000000000001453. American Heart Association guideline hub.
Sampson M, Ling C, Sun Q, et al. A New Equation for Calculation of Low-Density Lipoprotein Cholesterol in Patients With Normolipidemia and/or Hypertriglyceridemia. JAMA Cardiology. 2020;5(5):540-548. doi:10.1001/jamacardio.2020.0013.
McQueen MJ, Hawken S, Wang X, et al. Lipids, lipoproteins, and apolipoproteins as risk markers of myocardial infarction in 52 countries (the INTERHEART study): a case-control study. Lancet. 2008;372(9634):224-233. doi:10.1016/S0140-6736(08)61076-4.
Chen Y, Chang Z, Liu Y, et al. Triglyceride to high-density lipoprotein cholesterol ratio and cardiovascular events in the general population: a systematic review and meta-analysis of cohort studies. Nutr Metab Cardiovasc Dis. 2022;32(2):318-329. doi:10.1016/j.numecd.2021.11.005. PMID:34953633.
Assempoor R, Daneshvar MS, Taghvaei A, et al. Atherogenic index of plasma and coronary artery disease: a systematic review and meta-analysis of observational studies. Cardiovasc Diabetol. 2025;24(1):35. doi:10.1186/s12933-025-02582-2. PMID:39844262.

Medical disclaimer: This article and the GlobalRPh Cardiovascular Risk and Advanced Lipid Analyzer are intended for education and clinical decision support. They do not diagnose cardiovascular disease, prescribe therapy, establish an individualized treatment target, or replace evaluation by a qualified healthcare professional. PREVENT produces population-derived risk estimates and should be applied only in appropriate populations. Laboratory methodology, medications, pregnancy, acute illness, prior untreated values, established ASCVD, subclinical atherosclerosis and other patient-specific factors can materially change interpretation.