hs-CRP: the inflammation marker that out-predicts LDL.

High-sensitivity C-reactive protein, written hs-CRP, measures a protein the liver releases when interleukin-6 tells it to. It is the same molecule a hospital measures to track infection, read on a more sensitive assay that resolves the low concentrations relevant to arteries rather than abscesses, roughly 0.2 to 10 mg/L. Below 1.0 mg/L is low cardiovascular risk. Above 3.0 mg/L is high. The test costs a few dollars and sits on most Australian pathology menus. It is ordered far less often than it should be.

What a liver protein knows about your arteries

Atherosclerosis was taught for decades as a plumbing problem. Cholesterol accumulates, the pipe narrows, blood stops. That model explains the anatomy and misses the biology. What converts a stable plaque into a heart attack is usually not gradual narrowing but rupture of a thin fibrous cap, and cap thinning is an inflammatory process driven by macrophages, interleukin-1β and interleukin-6 inside the arterial wall.

hs-CRP is the downstream echo of that signalling. The liver makes CRP in proportion to circulating IL-6, so a single cheap blood test gives a rough readout of how much inflammatory traffic the body is running. It says nothing about where the inflammation is. Arthritis, a chest infection, gum disease, visceral fat and an unstable coronary plaque all push the same number up, which is exactly why the test is misread so often.

The epidemiology is unusually consistent. The Emerging Risk Factors Collaboration pooled individual data on 160,309 people without vascular disease across 54 prospective studies and found a near log-linear relationship between CRP concentration and later coronary heart disease, ischaemic stroke and vascular death. The association survived adjustment for conventional risk factors, though it shrank considerably once cholesterol, blood pressure, smoking and diabetes were accounted for.

The same collaboration then asked the harder question in the New England Journal of Medicine in 2012: does adding hs-CRP to a conventional risk score change what happens to patients? Across 246,669 participants, adding CRP to standard risk factors in people at intermediate risk was estimated to prevent one additional cardiovascular event per 400 to 500 people screened over 10 years. Real, and modest. Anyone selling hs-CRP as a crystal ball is overselling a test that earns its place through a narrower and more interesting claim.

The claim that actually matters: residual inflammatory risk

Put a patient on a decent statin and their LDL cholesterol falls by half. Some of them still have heart attacks. For thirty years the assumed explanation was that the LDL had not fallen far enough, and the field responded by pushing LDL targets lower.

Ridker and colleagues tested that assumption directly in the Lancet in 2023, pooling 31,245 patients on contemporary statins from three large trials: PROMINENT, REDUCE-IT and STRENGTH. They compared quartiles of baseline hs-CRP against quartiles of baseline LDL cholesterol as predictors of what happened next. The results were not close.

• Major adverse cardiovascular events: highest versus lowest hs-CRP quartile, adjusted hazard ratio 1.31 (95% CI 1.20–1.43). For LDL cholesterol, 1.07 (0.98–1.17), which did not reach significance.
• Cardiovascular death: hs-CRP 2.68 (2.22–3.23). LDL cholesterol 1.27 (1.07–1.50).
• All-cause death: hs-CRP 2.42 (2.12–2.77). LDL cholesterol 1.16 (1.03–1.32).

Among people already taking a statin, the inflammatory marker predicted death roughly twice as strongly as the cholesterol marker, and the ranges and relationships were near identical across all three trials. That is a substantial finding about the population most cardiology clinics spend their time managing, and it has barely dented ordering habits in Australian general practice.

None of this demotes ApoB. Lowering apoB-containing particles remains the single best-evidenced way to prevent a first heart attack, and we argue that case in detail in why ApoB beats LDL cholesterol. The point is what remains after that job is done.

Four trials, two of them negative

Prediction is cheap. The question worth answering is whether treating inflammation changes outcomes, and the trial record here is genuinely mixed. It rewards reading in full rather than in headlines.

• JUPITER (Ridker et al. NEJM 2008): 17,802 apparently healthy adults with LDL below 3.4 mmol/L but hs-CRP at or above 2.0 mg/L, randomised to rosuvastatin 20 mg or placebo. Stopped early at a median 1.9 years. Hazard ratio 0.56 (95% CI 0.46–0.69) for first major cardiovascular events. Rosuvastatin cut LDL by 50% and hs-CRP by 37%, so the trial cannot separate the two effects. It also showed more new-onset diabetes in the statin arm.
• CANTOS (Ridker et al. NEJM 2017): 10,061 patients with prior myocardial infarction and hs-CRP at or above 2 mg/L, given canakinumab, a monoclonal antibody against interleukin-1β. At 150 mg the hazard ratio for recurrent events was 0.85 (0.74–0.98, p=0.021), with a 37% median reduction in hs-CRP and no change in lipids. First proof that lowering inflammation alone reduces cardiovascular events. Fatal infections were more common on the drug, and all-cause mortality did not improve.
• CIRT (Ridker et al. NEJM 2019): low-dose methotrexate in 4,786 patients with stable atherosclerosis plus diabetes or metabolic syndrome. No reduction in events. Critically, methotrexate did not lower IL-1β, IL-6 or CRP either. The negative result supports the pathway rather than refuting it: an anti-inflammatory that fails to move the relevant cytokines fails to move outcomes.
• LoDoCo2 (Nidorf et al. NEJM 2020): colchicine 0.5 mg daily in 5,522 patients with chronic coronary disease. Hazard ratio 0.69 (0.57–0.83) over a median 28.6 months, driven by fewer myocardial infarctions and revascularisations. All-cause mortality was not reduced, with a point estimate that went the wrong way.

Then CLEAR SYNERGY landed. Jolly and colleagues randomised 7,062 patients within 72 hours of PCI for myocardial infarction to colchicine or placebo and followed them for a median 3 years. Primary events occurred in 9.1% on colchicine and 9.3% on placebo. Flat. That result surprised a field that had spent five years treating colchicine as settled, and it is a useful corrective: the inflammatory hypothesis is real, but the drug, the dose, the patient and the timing all matter, and we do not yet know which combinations work.

The honest summary of inflammation as a drug target in 2026 is one clear positive trial with an expensive antibody and an infection signal, one positive trial with a cheap old drug that a larger trial failed to replicate, and one informative null. The honest summary of hs-CRP as a measurement is much stronger, because knowing a patient carries residual inflammatory risk changes how hard everything else gets managed.

Where the biology gets awkward

CRP is not the villain. It is the smoke alarm.

Zacho and colleagues, writing in the New England Journal of Medicine in 2008, used Mendelian randomisation across large Copenhagen cohorts: people who inherit gene variants that raise CRP for life do not have more ischaemic heart disease or stroke. Subsequent genetic analyses agreed. So a drug that lowered CRP and nothing else would be useless, and any clinic offering to "treat your CRP" has misunderstood the paper it is quoting.

Genetic studies of the interleukin-6 receptor point the other way, which is why IL-6 rather than CRP has become the pharmacological target of interest, and why IL-6 inhibitors are now in cardiovascular outcome trials. CRP remains the practical marker because it is stable in stored samples, standardised across laboratories, and costs a fraction of a cytokine assay.

Interpretation has one more trap worth naming. A result above 10 mg/L almost always reflects something acute: a virus, a flare, a recent injury, sometimes a hard training block. It does not describe baseline vascular risk and should be repeated after two weeks rather than acted on. Anyone reporting an hs-CRP of 14 mg/L as a cardiovascular finding without asking whether the patient had a cold that week is doing it wrong.

How to read your result

Every biomarker in The HodieLabs Evidence-Based Clinical Library is interpreted against published clinical authority standards. For hs-CRP the bands come from the AHA/CDC 2003 scientific statement, with the 2 mg/L threshold used by both JUPITER and CANTOS as the entry criterion for treatment trials.

• Below 1.0 mg/L, low risk. Little inflammatory contribution to your cardiovascular risk.
• 1.0–3.0 mg/L, average risk. The 2 mg/L mark inside this band is where the major trials drew their line.
• Above 3.0 mg/L, high risk. Worth confirming on a repeat sample, then explaining. Visceral adiposity, insulin resistance, fatty liver, untreated sleep apnoea, periodontal disease and autoimmune conditions cover most of the cases we see.
• Above 10 mg/L: treat as an acute-phase response, not a vascular result. Retest.

The value comes from pairing. An hs-CRP of 4 mg/L in someone with a waist of 105 cm, ApoB of 1.2 g/L and fasting insulin high enough to flag insulin resistance is a coherent metabolic picture with an obvious intervention. The same 4 mg/L in a lean, fit 42-year-old with clean lipids is a question, not an answer, and the next step is finding the source rather than adding a drug.

What actually moves the number

Weight loss does most of the work. In the Diabetes Prevention Program, one year of intensive lifestyle intervention lowered CRP by 33% in men and 29% in women, against roughly no change on placebo, and the analysis found weight loss rather than increased activity accounted for most of the effect. Metformin produced smaller reductions. That is a large effect for an intervention with no side-effect profile to speak of.

Statins lower hs-CRP by about a third, and they do it largely independently of how much they lower LDL, which is the pleiotropic effect JUPITER was built to exploit. Treating obstructive sleep apnoea, periodontal disease and metabolic dysfunction-associated steatotic liver disease lowers it where those conditions are driving it. Regular aerobic exercise lowers it, mostly through changes in visceral fat rather than any direct anti-inflammatory magic.

Nothing on the supplement shelf has convincing outcome data behind it. Omega-3 in the STRENGTH trial failed to reduce cardiovascular events despite lowering triglycerides, and curcumin, resveratrol and their neighbours have small, short, surrogate-endpoint trials and nothing more. A patient with an hs-CRP of 5 mg/L and a 105 cm waist is not short of turmeric.

The Australian gap

Cardiovascular disease still contributes to roughly one in four Australian deaths, and the standard preventative workup here is a lipid panel, a blood pressure reading and an absolute risk calculator. hs-CRP is not in the routine panel, is not in the Australian absolute cardiovascular disease risk calculator, and has to be requested by name. So the marker that best predicts who dies among people already on a statin is the one almost nobody in the country is measuring.

We measure hs-CRP in every Health Blueprint at our Melbourne Preventative Health & Longevity Clinic, alongside ApoB, Lp(a), fibrinogen, fasting insulin and body composition. It is read as part of a pattern by a doctor, repeated when it looks acute, and used to decide how aggressively the modifiable risks get managed. On its own it is a number. In context it is one of the more useful numbers on the panel.

If you are on a statin, your LDL is at target and you have never had your hs-CRP measured, then the biggest remaining piece of your cardiovascular risk has never been looked at. Ask for it by name at your next blood test, and if it comes back above 2 mg/L, ask why.

References

1. Ridker PM, et al. Rosuvastatin to prevent vascular events in men and women with elevated C-reactive protein (JUPITER). N Engl J Med. 2008;359:2195–2207.
2. Ridker PM, et al. Antiinflammatory therapy with canakinumab for atherosclerotic disease (CANTOS). N Engl J Med. 2017;377:1119–1131.
3. Ridker PM, et al. Low-dose methotrexate for the prevention of atherosclerotic events (CIRT). N Engl J Med. 2019;380:752–762.
4. Nidorf SM, et al. Colchicine in patients with chronic coronary disease (LoDoCo2). N Engl J Med. 2020;383:1838–1847.
5. Jolly SS, et al. Colchicine in acute myocardial infarction (CLEAR SYNERGY). N Engl J Med. 2025;392:633–642.
6. Ridker PM, Bhatt DL, Pradhan AD, et al. Inflammation and cholesterol as predictors of cardiovascular events among patients receiving statin therapy: a collaborative analysis of three randomised trials. Lancet. 2023;401:1293–1301.
7. Emerging Risk Factors Collaboration. C-reactive protein concentration and risk of coronary heart disease, stroke, and mortality: an individual participant meta-analysis. Lancet. 2010;375:132–140.
8. Kaptoge S, et al. C-reactive protein, fibrinogen, and cardiovascular disease prediction. N Engl J Med. 2012;367:1310–1320.
9. Zacho J, et al. Genetically elevated C-reactive protein and ischemic vascular disease. N Engl J Med. 2008;359:1897–1908.
10. Haffner S, et al. Intensive lifestyle intervention or metformin on inflammation and coagulation in participants with impaired glucose tolerance. Diabetes. 2005;54:1566–1572.
11. Pearson TA, et al. Markers of inflammation and cardiovascular disease: application to clinical and public health practice. AHA/CDC scientific statement. Circulation. 2003;107:499–511.

This article is general information, not medical advice. Speak with your GP about what testing and care is appropriate for you.