Measuring Silent Inflammation: How to Read hs-CRP and Other Blood Tests

You are holding your lab report. CRP: unremarkable. A tick, a green box, all good.

And yet in the evening you read about silent inflammation everywhere. About a fire smouldering quietly in the body. About a value that measures more sensitively. About panels with twelve markers.

Many people know this pattern. The tiredness is there, the belly has become a little rounder, sleep is not what it used to be. And somewhere between the lab sheet and the search engine the question arises: is something burning that nobody can see?

I don't want to scare you and I don't want to sell you a test. I want to show you what these values measure, when a measurement misleads and what the big trials have shown, including where they disappointed.

Silent inflammation: what is meant and how big the difference to acute inflammation is

Imagine two sounds. One is a fire engine siren right outside your window. The other is a smoke alarm in the next room that briefly beeps once every few minutes.

Both are alarm signals. But they ask something completely different of you.

The siren is acute inflammation. You cut your finger, you get pneumonia, a tooth abscess forms. The area turns red, warm, swollen and painful, and fever may follow. The immune system mobilizes everything it has. And then, if all goes well, it winds down again.

The smoke alarm is what is meant by silent inflammation. No fever, no redness, no clear location. Just an immune system that keeps running at a low level even though no acute trigger is recognizable anymore.

How CRP is made: a phone call and a loudspeaker announcement

C-reactive protein, CRP for short, is made in the liver. The order to make it comes mainly from a messenger of the immune system, interleukin-6.

Think of IL-6 as a phone call made by an immune cell somewhere in the body. CRP is the loudspeaker announcement the liver then sends through the entire bloodstream. The announcement says that something is going on. It does not say where.

After an acute stimulus, CRP rises above 5 mg/L after about six hours, peaks around 48 hours and has a half-life of about 19 hours. So it closely follows what is happening right now. That makes it useful and at the same time prone to snapshots.

These numbers are the key to everything that follows. When people talk about silent inflammation, they usually mean values somewhere between roughly 1 and 10 mg/L. Pneumonia can reach a hundred times that. These are not two steps of the same staircase, they are two different buildings.

How fat tissue and inflammation influence each other is covered in Silent inflammation and weight. How inflammation, the gut and energy metabolism might interact in exhaustion is discussed in Burnout, gut, inflammation and mitochondria.

And now you know why the first question about an inflammation marker should be: what order of magnitude are we actually talking about?

hs-CRP in detail: a finer scale, three risk tiers and the reason for two measurements

Some reports say CRP, others hs-CRP. Sounds like two different substances. It isn't.

The same protein, a finer scale

Imagine a bathroom scale and a letter scale. With the bathroom scale you can easily tell whether a suitcase weighs 20 or 23 kilos. Whether a letter weighs 12 or 18 grams, it won't tell you.

Standard CRP is the bathroom scale, built for acute inflammation with values of ten, fifty or two hundred mg/L. Where its lower detection limit lies depends on the lab and the method.

hs-CRP is the letter scale. hs stands for high sensitivity. It measures the same molecule with a method that still distinguishes reliably in the range below 1 mg/L. That is exactly the range that matters when it comes to silent inflammation.

How finely this letter scale measures was described by Roberts in a background paper for the US Centers for Disease Control and Prevention (CDC) and the American Heart Association: many high-sensitivity methods reach an imprecision of under 7 percent at values below 1 mg/L. Cytokines and related molecules, by contrast, are not measured routinely in clinical labs. That is one reason why hs-CRP appears in so many studies on silent inflammation.

Check the unit. Some labs report CRP in mg/L, others in mg/dL. There is a factor of 10 between them. A value of 0.3 mg/dL equals 3 mg/L. Overlook the unit and you can easily end up a whole risk tier off.

You don't need to fast for hs-CRP. According to Pepys and Hirschfield, CRP shows no daily rhythm and is not affected by food. If blood glucose or blood lipids are measured at the same appointment, fasting may be needed for those values. The practice drawing your blood will decide that.

And what does it cost? There is no reliable single figure. Whether the test is covered depends on the reason for testing and on your insurer.

You deliberately won't find a table of normal values here. Labs work with different methods, units and comparison groups, and they print their own reference range on the report. A general table would pretend there is one single correct number.

Where the tiers below 1, 1 to 3 and above 3 mg/L come from

Almost every page on the topic lists these three tiers. Hardly any say where they come from and what they were meant for.

The 2019 US guideline on primary prevention uses a different threshold: hs-CRP of 2.0 mg/L or more as one of several risk-enhancing factors, explicitly only if it has been measured.

Why a single measurement can put you in the wrong box

Here comes the number that impressed me most during my research.

Even in healthy adults in ordinary daily life, the value moves. Ockene and colleagues had hs-CRP and total cholesterol measured five times over one year in 113 healthy adults. The variability was comparable for both. When divided into quartiles, the first and second hs-CRP measurements fell into the same quartile in 63 percent of cases, for cholesterol in 60 percent. So not a random value, but a number that can place you one tier too high or too low.

That is why Pepys and Hirschfield recommend repeating higher values at least one week apart, the AHA and CDC about two weeks apart. The point is the same: don't measure once and then decide.

What shifts your value temporarily

CRP is a non-specific response. The liver doesn't ask where the call is coming from. That is why in some situations a value comes out higher than it would be in your baseline state.

• Infections and injuries. Any acute inflammation, even an infection you barely noticed, can raise CRP.
• Intense endurance exercise. In the short term, exertion is an inflammatory stimulus, and a study on this follows shortly.
• Gums and dental treatment. In cross-sectional studies, periodontitis was associated with a CRP that was 1.56 mg/L higher, and directly after intensive periodontal therapy CRP was significantly raised at 24 hours. More on this below.
• Hormone preparations. According to Pepys and Hirschfield, the pill and oral hormone replacement therapy during menopause, but not therapy given through the skin, are associated with higher baseline CRP without any recognizable tissue-damaging inflammation behind it. That is a reason to mention the medication before the blood draw, not a reason to stop it. Any change belongs in the hands of the prescribing doctor.

That does not mean exercise fuels inflammation. The systematic review by Kasapis and Thompson sums it up like this: exertion produces a short-term inflammatory response, while training studies and cross-sectional comparisons show an anti-inflammatory effect in the long term. More on how the body recovers after hard sessions can be found in Recovery after exercise.

And now you know why the second measurement is not mistrust of the lab, but the precondition for the first one meaning anything at all.

JUPITER and CANTOS: can less inflammation prevent heart attacks?

Imagine the smoke alarm in your hallway has been beeping for weeks. You have two options. You can take out the battery. Or you can go and look at what is smouldering.

That is exactly the question behind the big inflammation trials in cardiology: is CRP itself part of the problem, or does it only indicate something?

Smoke alarm or fire? What genetics says

Some people are born with gene variants that keep their CRP slightly higher throughout life. If CRP itself helped cause heart attacks, they would have to be affected more often. This natural experiment is called Mendelian randomization.

One level higher, things look different. The IL6R consortium around Swerdlow examined, across 40 studies, a gene variant that dampens the IL-6 signal much like a receptor blocker. It was associated with 8.35 percent less CRP per allele and slightly fewer coronary events, odds ratio 0.95. So the signalling pathway that triggers CRP appears to be involved. CRP itself probably only indicates it.

JUPITER: normal LDL, raised hs-CRP

Now the catches, and they belong here just as much.

• Early stopping, small absolute rates. Trials stopped early can make effects look larger. And 0.77 versus 1.36 events per 100 person-years is a clear difference in relative terms. What it means in absolute terms for an individual depends on their baseline risk and belongs in a medical consultation.
• LDL was halved. The benefit therefore cannot be attributed to lowering inflammation alone. The 2019 US guideline even writes about JUPITER that the magnitude of percentage LDL lowering determined the benefit.
• No comparison arm with low hs-CRP. Because only people with 2 mg/L or more were enrolled, JUPITER does not show whether people with low hs-CRP would have benefited just as much.
• Diabetes. In a secondary analysis, the diabetes rate rose by 28 percent in people with at least one diabetes risk factor, while 134 vascular events or deaths were avoided for every 54 new diabetes diagnoses. Without a risk factor, there was no rise in diabetes. JUPITER was funded by AstraZeneca.
• Published criticism. A group of authors around de Lorgeril considered the trial methodologically flawed and criticized the early stopping and commercial influence. The guidelines included JUPITER despite this criticism.

One sentence so that nothing here is misunderstood: If you take a statin, this article changes nothing about that. Whether a statin makes sense depends on your overall risk and belongs in a medical consultation, not in a blog article. How such an indication comes about is explained in Statins from 35? Why high LDL alone is not a diagnosis.

CANTOS: lowering inflammation without touching cholesterol

JUPITER could not answer the question cleanly because the statin did two things at once. CANTOS therefore took an approach away from cholesterol.

A prespecified secondary analysis went one step further. Those who reached an hs-CRP below 2 mg/L on canakinumab had 25 percent fewer major cardiovascular events, hazard ratio 0.75. Those who stayed at 2 mg/L or above did not benefit significantly, hazard ratio 0.90. The catch: these groups only formed after randomization, and people who respond well may also differ in other respects. The authors adjusted for this, but bias cannot be ruled out entirely.

Canakinumab is not authorized in the European Union for cardiovascular prevention. The manufacturer withdrew its application for this use at the European Medicines Agency on 4 December 2018. According to the provisional view of the responsible committee at that time, the data were not sufficient to clearly show efficacy in all people after a heart attack. The benefit was classed as modest, especially in people already taking a statin, and was not judged to outweigh the higher risk of serious infections. The committee also questioned whether the chosen measure was suitable for selecting patients and monitoring efficacy.

The trials that disappointed: not every anti-inflammatory approach pays off

If only CANTOS existed, the story would be simple. But two of three further large trials don't fit the neat picture.

In CIRT, led by Ridker, events were 4.13 per 100 person-years on methotrexate versus 4.31 on placebo. Unlike CANTOS, a raised hs-CRP is not named as an inclusion criterion there, so perhaps many participants lacked the target to begin with. In LoDoCo2, led by Nidorf, colchicine lowered events in stable coronary heart disease, while in CLEAR, led by Jolly, directly after a heart attack it did not, even though CRP fell.

CLEAR in particular is instructive. The lab value went down, the cardiovascular events stayed the same. A lower value is therefore not automatically a better outcome.

A pooled analysis by Ridker and colleagues of 31,245 people on statin therapy from three trials adds to the picture. Comparing the top with the bottom quartile, the hazard ratio for major cardiovascular events was 1.31 for high hs-CRP and 1.07 for high LDL, which was not significant. On statins, hs-CRP therefore predicted more than LDL here. This is an observational analysis in people at high risk, funded by the manufacturers of the drugs tested there, and explicitly not an argument against ongoing statin therapy.

And now you know why this article doesn't give you instructions for pushing a value down with pills, but for asking about the cause.

Measure or not? Why the guidelines contradict each other

You ask your GP practice about hs-CRP and hear: we don't need that for your risk calculation. In the evening you read that a large US professional society wants this very value measured widely. Who is right?

The honest answer: both sides have good reasons. And they are not answering quite the same question.

The Emerging Risk Factors Collaboration around Kaptoge reports the same figure for its analysis of 246,669 participants from 52 studies: CRP increased the C-index by 0.0039. Both figures very likely rest on the same data, so they are not two independent confirmations. According to their model, targeted measurement in people at intermediate risk could prevent one additional event per 400 to 500 people screened over ten years. Small, but not zero.

The US side reads the same data differently. In the 2019 ACC and AHA guideline, a measured hs-CRP of 2.0 mg/L or more counts among the risk-enhancing factors. At a ten-year risk of 7.5 to under 20 percent, such factors favor starting or intensifying statin therapy within the clinician-patient risk discussion (class of recommendation IIa, level of evidence B-R). At a borderline risk of 5 to under 7.5 percent, they may justify moderate-intensity statin therapy within that discussion (class IIb, level of evidence B-R).

In 2025, the American College of Cardiology went further. Its position paper states that residual inflammation is strongly associated with further events even on statins and, according to the official summary, recommends broad screening for hs-CRP together with LDL. But it also warns that not all anti-inflammatory trials in secondary prevention were successful.

And laboratory medicine? In 2009, a panel of the US National Academy of Clinical Biochemistry led by Myers examined many newer markers, from fibrinogen and homocysteine to kidney values. Only hs-CRP met all criteria for a biomarker for risk assessment.

The question no guideline answers

All these papers ask whether hs-CRP improves the calculation for heart attack and stroke. Functional medicine asks a different question alongside it: is a repeatedly raised value a reason to look for causes, such as periodontitis, a sleep disorder, visceral fat or a previously unrecognized inflammatory disease?

To be honest: the guidelines reviewed here don't answer it, and I have not found a study that demonstrates the benefit of this approach with hard endpoints. It is clinically understandable, not proven.

And now you know why your GP and a US cardiologist can reach different recommendations for the same lab value without either of them doing a bad job.

Other blood markers: ESR, ferritin, IL-6, NLR, omega-3 index and what inflammation panels can't deliver

Maybe you have already seen an offer like this: an inflammation check with twelve markers and traffic-light colours. That feels more thorough than a single value. But every marker has its own strength and its own weakness.

Where ferritin in the blood comes from during inflammation has not been fully clarified. Kell and Pretorius suggest that it leaks mainly from damaged cells, a mechanistically plausible hypothesis, not an established finding. More in Ferritin: what is normal, Iron and inflammation and Iron deficiency: which blood values count.

And the big inflammation panels?

Many self-pay packages combine cytokines, LPS antibodies, zonulin, fatty acid profiles and more. Three findings put this into context.

• The team around Furman writes that there are currently no standard biomarkers for harmful chronic inflammation.
• According to Roberts, cytokines are not measured routinely in clinical labs.
• The panel led by Myers found that among many newer markers only hs-CRP met all criteria.

That does not mean these panels are worthless. For most of their markers, however, there is no validated use for diagnosing silent inflammation in an individual person. What lies behind zonulin is discussed in Leaky gut and zonulin.

And now you know why an hs-CRP measured properly twice, together with a blood count and ferritin, can often tell you more than a package of twelve markers.

Where silent inflammation comes from: the causes and their evidence

When the smoke alarm beeps, the next question is not how loud it is. It is which room you should check.

From the perspective of Clinical Psychoneuroimmunology, KPNI for short, four lenses are worth using at once: the immune system, metabolism, the nervous system and the hormonal system. From this angle, silent inflammation is rarely a matter of the immune system alone. It can arise where these systems nudge each other.

Visceral fat: the fat around the organs

How fat tissue and inflammation can influence each other is covered in Silent inflammation and weight.

Physical inactivity

Kasapis and Thompson found a long-term anti-inflammatory effect of physical activity, and a meta-analysis in the next section shows CRP lowering through exercise even without weight loss. Why daily stillness matters more than the hours of sport is explained in Sitting and physical inactivity.

Sleep

This is where the nervous system lens comes in: sleep is regarded as the phase in which the stress axis and the immune system rebalance. If you snore loudly, gasp for air at night or wake up feeling wrecked, sleep apnoea may be behind it. How to recognize it is covered in Recognizing sleep apnoea.

Smoking

In two Italian lung cancer screening studies with 3,050 heavy smokers and ex-smokers, Gallus and colleagues found a high CRP of 2 mg/L or more in 35.8 percent of ex-smokers and in 41.1 percent of current smokers. After more than four years without smoking, CRP fell with each additional year, and at more than eight years the odds ratio was 0.55. In the longitudinal analysis over an average of 3.4 years, by contrast, there was no significant decrease. The authors write that the benefit is not visible in the short term.

In a cohort led by King with 1,652 smokers, smoking intensity was mainly related to leukocytes and oxidative stress, not to hs-CRP. After one year of quitting, leukocytes and oxidative stress fell, but hs-CRP did not.

Periodontitis: a look inside the mouth

Autoimmune processes and chronic infections

The team around Furman lists chronic infections among the triggers, and inflammatory rheumatic diseases such as rheumatoid arthritis are often accompanied by persistently raised CRP. If a raised CRP persists and swollen joints, fever, weight loss or night sweats come along with it, this needs prompt medical assessment. How autoimmune processes can develop, using the thyroid as an example, is described in Hashimoto's thyroiditis: causes in the immune system.

Eating patterns

Diet appears in Furman's list among the possible triggers. Which foods tend to be regarded as pro-inflammatory is covered in Pro-inflammatory foods: fries or banana. What an eating pattern changed in hs-CRP in studies is covered in the next section.

And your genes

Part of your baseline CRP may simply be inherited. According to Pepys and Hirschfield, twin studies show a clear hereditary component. In the CCGC data, individual gene variants were associated with up to 30 percent difference in concentration per allele, and people with such variants did not have a higher heart risk because of them.

And now you know why asking about the cause yields so much more than asking how high the value is.

What lowered hs-CRP in studies, with effect sizes

Maybe this question has been on your mind from the start: and how do I get the value down?

After everything genetics and CLEAR have shown, the goal is not the smaller number, but less of whatever drives it up. The good news: the measures that lowered hs-CRP in studies are almost all levers of that kind, aimed at causes.

Weight: a study figure, not a target

Selvin, Paynter and Erlinger pooled 33 intervention studies ranging from lifestyle to surgery. Across all studies, CRP fell by an average of 0.13 mg/L per kilogram of weight lost, weighted correlation 0.85, and only 0.30 in pure lifestyle studies. My interpretation, not the authors': the strong values probably come in part from the surgical studies.

Exercise: not only through the scales

What exercise can set in motion at the cellular level is described in Exercise as medicine at the cellular level. And why muscle strength becomes more important over the years is covered in Strength training after 40.

A Mediterranean eating pattern

In a well-known single trial by Esposito and colleagues, 180 people with metabolic syndrome ate a Mediterranean or a low-fat diet for two years. Afterwards, hs-CRP and IL-6 were significantly lower in the Mediterranean group. However, that group also lost more weight, so diet and weight loss cannot be cleanly separated.

The DEGAM guideline also recommends a way of eating based on the Mediterranean diet, regardless of the level of risk. Here the functional perspective and the guideline agree.

Gums: a double-edged sword for measurement

In the meta-analysis by Teeuw and colleagues, hs-CRP fell by an average of 0.50 mg/L after periodontal treatment. Whether this leads to fewer heart attacks was not measured by any of the 25 studies.

Quitting smoking, turmeric and medications

After quitting smoking, CRP takes years, not weeks. If you don't see a lower value three months after your last cigarette, you haven't done anything wrong.

Turmeric is often mentioned as a natural anti-inflammatory. What the studies on curcumin actually show and where the hype goes beyond the data is covered in Turmeric for inflammation: what curcumin can do.

Medications that lowered inflammation markers in studies are deliberately not in this table. They have their place in the medical treatment of specific diseases, after medical consideration and prescription, and they are not a means of improving a lab value.

And now you know why the measures with the best data are not special anti-inflammatory remedies, but perfectly ordinary basics.

The limit: a clue, not a diagnosis

To finish, two sentences that easily get lost amid all the enthusiasm for lab values.

First: a raised value is non-specific. The CRP response says that something is going on, not what and not where. Pepys and Hirschfield explicitly call it non-specific, and Bray and colleagues stress that ESR and CRP should only be used together with the medical history and examination. A raised hs-CRP is a clue that opens a search. It is not a diagnosis that ends it.

Second: a normal value rules nothing out. The textbook example is systemic lupus erythematosus. In this autoimmune disease, a substantial CRP response is often absent even though inflammation and tissue damage are clearly present. ESR and CRP also often diverge, as described above. If you have symptoms, those symptoms count more than a green box on the report.

And a third sentence follows from CLEAR: a lowered value is not automatically a better outcome.

That sounds less exciting than an inflammation panel with traffic-light colours. But it leaves you with something no test can give you: the freedom not to chase every number, and the clarity to look closely where it is worth it.

If you take away only three things

• Don't have it measured once, have it measured twice. About two weeks apart, during a phase without an infection, without a competition and without recent dental treatment. Tell the practice whether you take hormone preparations.
• Look first where the studies point to causes. Gums, sleep, movement in daily life, smoking and the fat around the organs. That is unspectacular and often more rewarding than another lab package.
• Don't change anything about your medications because of a value. Not a statin, not cortisone, hormones or painkillers. That belongs in the hands of the doctor who prescribed them.

You can find more articles on nutrition and inflammation in the nutrition topic hub. And if you would like not just to read but to put your values into context together: below this article you will find the option to book an appointment.

And now you know why a single number never has the last word, neither for better nor for worse.

Frequently asked questions about silent inflammation and hs-CRP

Where to go from here

Silent inflammation touches on weight, iron, fatty acids, nutrition, sleep and exercise. Each of these articles explores one of the threads that are only touched on here.