hs-CRP: The Inflammation Workhorse

hs-CRP is the cheapest useful window into whether your body is running a low-grade inflammatory process. It is also the marker most often over-read, because a single value tells you almost nothing and a repeated value tells you a great deal. Understanding the difference — and understanding what CRP is actually reporting on — is what turns it from a number on a page into something you can act on.

What CRP is, and the pathway that produces it

C-reactive protein is made in the liver, and it is made on instruction. The signal is interleukin-6, released by immune cells and, importantly, by adipose tissue. When IL-6 reaches the liver, hepatocytes ramp up production of a set of acute-phase proteins, of which CRP is the most useful clinically (Pepys & Hirschfield, J Clin Invest 2003).

Two properties follow from that and explain most of its behaviour. It rises fast — within hours of an inflammatory stimulus, with a very large dynamic range — and it falls fast once the stimulus resolves, because its half-life is short and production stops as soon as the IL-6 signal does. It is therefore a live readout, not an average. It tells you what is happening now, not what happened last month.

CRP is not simply a bystander. It binds damaged cell membranes and certain bacterial surfaces and flags them for clearance. That is its actual job; the risk-marker use is a by-product of our being able to measure it.

"High sensitivity" is the assay, not the molecule

There is one CRP. The distinction is entirely in the measurement.

• Standard CRP was designed to detect obvious inflammation — infection, flare, serious illness — and reports across a wide range where the interesting values are large.
• hs-CRP uses a more sensitive method to resolve the bottom of that range, where the differences between 0.4, 1.2 and 2.8 mg/L are meaningful.

This matters when ordering. A standard CRP returned as "<5, normal" has answered a different question and cannot be used for risk stratification. If cardiovascular or metabolic risk is the reason for testing, the request has to specify high-sensitivity.

The ranges, and where they came from

• Below 1.0 mg/L — low risk
• 1.0 to 3.0 mg/L — intermediate
• Above 3.0 mg/L — high risk
• Above 10 mg/L — not a risk category at all. This is acute inflammation. Recheck once recovered.

These cut-points come from population risk stratification, not from a diagnosis. A value of 2.6 does not mean you have a disease; it means you sit in a band of the population with more events over time than the band below. That distinction should shape how much weight the number carries in any single person's decision.

Why one reading is not enough

This is the most consequential practical point on the page. Within-person variability in hs-CRP is substantial — repeat measurements in the same stable individual can land in different risk categories, and classification improves materially when two or more readings are averaged (Ockene et al., Clin Chem 2001).

The things that raise a reading transiently are mundane: a cold two days ago, a dental cleaning, a heavy training session, a minor injury, a flaring joint. None relate to cardiovascular risk, and all can push a value across a cut-point.

So the rule is simple. Two readings, at least two weeks apart, neither taken within a fortnight of an illness or injury. If they disagree wildly, the higher one is usually the artefact. Anyone acting on a single hs-CRP is acting on noise as often as on signal.

What raises it, and by what route

• Visceral fat. The most important entry on this list. Visceral adipose tissue secretes IL-6 directly into the portal circulation, so it reaches the liver at concentration — which is why waist circumference tracks CRP more closely than body weight does (Fontana et al., Diabetes 2007)
• Insulin resistance and metabolic syndrome, which travel with the above
• Smoking
• Poor sleep and untreated sleep apnoea, through intermittent hypoxia and sympathetic activation
• Periodontal disease — a genuinely under-appreciated chronic inflammatory source; treating it lowers systemic CRP (D’Aiuto et al., J Dent Res)
• Chronic psychological stress
• Autoimmune conditions, subclinical infection, and recent injury or surgery
• Oral estrogen, which raises CRP through first-pass hepatic effects without necessarily indicating inflammation elsewhere — a reading in this context needs care

What it adds beyond a lipid panel

The reason hs-CRP earned its place is that inflammatory risk and lipid risk are not the same axis. Long-term follow-up in women showed inflammation, cholesterol and Lp(a) each contributing independently to cardiovascular outcomes across thirty years (Ridker et al., N Engl J Med 2024). A person with good lipids and persistently raised CRP is not low risk, and treating them as such misses the point of measuring it.

The stronger evidence goes further than association. Trial data showed that treating people selected on raised CRP with normal cholesterol reduced events (Ridker et al., N Engl J Med 2008), and a drug that lowers inflammation without touching lipids at all reduced cardiovascular events (Ridker et al., N Engl J Med 2017). That second result is the one that established inflammation as a target rather than a signpost — though the drug in question is not a general-purpose intervention and came with its own costs.

In practice hs-CRP sits alongside ApoB rather than competing with it, and both belong in the panel described in inflammation markers explained. Fibrinogen adds a slower-moving second view of the same process.

What lowers it, and how fast

Nothing on this list acts on CRP directly. Each removes an upstream driver, and CRP falls because the IL-6 signal falls.

• Losing visceral fat — the largest lever available to most people
• Regular aerobic and resistance exercise, though a hard session raises it acutely, so time the draw
• Treating sleep apnoea and improving sleep duration
• Stopping smoking
• A Mediterranean-pattern diet
• Statins, which lower CRP partly independently of their lipid effect
• GLP-1 therapy, where clinically appropriate — see GLP-1 and inflammation
• Dental treatment, where periodontal disease is present

Timeline: expect three to six months before a retest means anything. CRP responds quickly to acute change but the chronic baseline moves at the pace of the tissue driving it, and visceral fat does not disappear in a fortnight.

What it cannot tell you

hs-CRP is entirely non-specific. It says inflammation is present somewhere; it says nothing at all about where or why. A raised value in someone with a large waist, poor sleep and untreated gum disease has three plausible explanations and no way of distinguishing between them from the number alone.

It is also not a screening test for occult disease in a well person, and chasing a mildly raised value with broad imaging tends to generate incidental findings rather than answers. The productive move is to work through the common, modifiable causes in order — waist, sleep, teeth, smoking, glucose — and only then to widen the search. A proper baseline panel gives that work somewhere to start.

Bottom line

hs-CRP measures a liver protein made in response to IL-6, which makes it a fast, sensitive and completely non-specific readout of systemic inflammation. Below 1.0 mg/L is the low-risk band, above 3.0 the high one, above 10 acute inflammation rather than risk. Its value is real and independent of lipids — but only when measured properly: twice, away from illness, with a high-sensitivity assay. Treat what is raising it, and give any intervention three to six months before retesting.

Educational content, not medical advice. Laboratory interpretation and any treatment decision are made by a licensed physician after individual evaluation. Individual results vary.