hs-CRP

What the research found.

• Two databases recorded what rheumatoid arthritis patients looked like at their first measurement. A normal CRP, under 10 milligrams per liter, turned up in 44 percent of one group and 58 percent of the other. Both inflammation markers read normal in 33 to 42 percent.

  Sokka and Pincus, Journal of Rheumatology, 2009

• Treating rheumatoid arthritis toward a combined score got more people into a quiet spell than ordinary care did, and that score includes an inflammation marker. The biggest trial, called TICORA, reported 65 percent against 16 percent. Several other trials pointed in that direction with smaller margins.

  Hock and colleagues, SN Comprehensive Clinical Medicine, 2021

• For spotting giant cell arteritis, C-reactive protein caught 74 percent of cases and cleared 49 percent of the people who didn't have it. The sedimentation rate caught 48 percent and cleared 72 percent. Neither one is good enough on its own, which is why both usually get ordered together.

  Currier and colleagues, Rheumatology International, 2025

• In that series, 23 of the 130 giant cell arteritis patients had two normal results at once, with the sedimentation rate and the C-reactive protein both reading normal. That's close to one patient in six. Every one of them had a disease that can take somebody's sight within days.

  Currier and colleagues, Rheumatology International, 2025

What the test measures

Your liver makes C-reactive protein when it gets inflammation signals, and the main signal is a messenger called interleukin-6. When inflammation starts anywhere in the body, the liver steps up production within hours, and when the inflammation settles, production falls just as fast. That quickness is what makes the test useful, and it's also what makes it easy to misread.

The protein does a job in immune defense, sticking to damaged cells and to some bacteria and helping mark them for clearing. For clinical purposes it works as a signal, not a mechanism. It tells you your liver is getting inflammation messages, and it doesn't tell you where those messages are coming from.

Why a normal result settles nothing

This is the finding worth taking away, because getting it wrong costs people years. Two databases of rheumatoid arthritis patients recorded their markers at first presentation, covering 1980 to 2004, one in Finland and one in the United States. A normal CRP, meaning under 10 milligrams per liter, turned up in 44 percent of patients in one group and 58 percent in the other.

Both markers read normal in 33 and 42 percent, and all three of the sedimentation rate, CRP, and rheumatoid factor read normal in 15 and 14 percent. All three read abnormal in only 28 and 23 percent. So the most common finding at presentation wasn't a full set of abnormal tests, and a large minority of people with real rheumatoid arthritis had entirely unremarkable bloods.

Suppose your joints are swollen and stiff and your markers are normal. That combination needs a rheumatology assessment rather than a discharge, because being sent away on the strength of normal markers is a known route to a late diagnosis, and delay in inflammatory arthritis costs joints. So say plainly that your joints are swollen, because that finding is what should drive the referral.

How it differs from the sedimentation rate

The two markers answer slightly different questions, and it helps to know which one you're looking at. CRP responds within hours in both directions, so it shows what's happening now, while the sedimentation rate depends on blood proteins that change over days to weeks. So the sedimentation rate shows the recent past, and it trails behind how you are.

The sedimentation rate is also affected by things that have nothing to do with inflammation, including your age, your sex, your blood counts, and pregnancy. The usual age rule for its upper limit has two forms, where for men you take your age and halve it, and for women you add ten to your age first and then halve it. CRP needs no such rule, which is one of its practical advantages.

In giant cell arteritis the two perform measurably differently. CRP caught 74 percent of cases and cleared 49 percent of healthy people, while the sedimentation rate caught 48 percent and cleared 72 percent. Neither is good enough alone, which is why both usually get ordered together.

Where the number does guide treatment

CRP on its own has no target, and it does real work inside a combined score. Modern rheumatoid arthritis care uses scores that pull several things into one number, those being joint counts, what you say, what your doctor sees, and an inflammation marker. Treatment then gets stepped up until that number reaches a set target, so CRP is one input to the sum rather than the thing being aimed at.

That approach has trial support behind it. One review pooled trials of following a protocol against ordinary care, and the protocol arms got far more people into a quiet spell, with the biggest difference coming from a trial called TICORA at 65 percent against 16 percent. Several other trials pointed in that direction with smaller margins.

Note what the evidence is for, though, because it's easy to overread. It's for the approach and for the combined score, and it isn't for any CRP number by itself. No trial has ever aimed at a CRP value alone.

What else moves it

Almost any inflammation raises CRP, which is why you can't read it alone. Infection is the most important other explanation, and a bacterial infection often pushes it far higher than arthritis does. An injury raises it, recent surgery raises it, and extra body weight raises it a little as well.

That counts in practice. A rising CRP in somebody with rheumatoid arthritis raises a question rather than answering one, because a flare and an infection can look alike and need opposite treatment. One calls for more immune-damping medicine and the other for less, so working out which you have takes an examination, a history, and sometimes more tests.

So report a rise to your team rather than reading it yourself. That isn't a counsel of helplessness, it is the one situation on this page where guessing has a real cost attached to it. Getting it backwards makes somebody sicker.

The high-sensitivity form

The test labeled hs-CRP measures the same protein by a more sensitive method, and it can tell apart values in the low normal range. It was built for judging heart risk, where the differences that count are small and an ordinary test would miss them. Rheumatology picked it up afterward rather than driving its development.

In active inflammatory arthritis your levels usually sit well above that range, so the ordinary test generally tells you what you need. Where the sensitive form earns its place in rheumatology is heart risk, and inflammatory arthritis raises that risk on its own, whatever your joints are doing. That rise is substantial and often ignored.

So if you've had hs-CRP measured, ask which question it was answering. The answer changes how to read it. A number ordered for your heart and a number ordered for your joints are two different numbers wearing one name.

Giant cell arteritis, where this gets urgent

One situation deserves separating out, because the stakes are higher than anywhere else on this page. Giant cell arteritis can take your sight for good within days, and normal inflammation markers don't rule it out. In one series of 130 patients with the diagnosis, 23 had a normal sedimentation rate and a normal CRP together, and the ones measured longer after their symptoms began were more likely to look normal.

So four things need urgent care in anyone over fifty, whatever the bloods show: a new bad headache, pain in the jaw when you chew, a tender scalp, and any change in your vision. That's the one place on this site where the advice is to act on your symptoms and let the tests catch up. Sight lost to this illness doesn't come back.

Questions patients ask.