Inflammation Markers: hs-CRP, Homocysteine, and What to Do

Two of the most informative labs available are also two of the least frequently ordered. hs-CRP and homocysteine both rise years before the events they are associated with, both are cheap, and neither appears on a standard annual panel. They are also both routinely over-interpreted, which is the other half of the problem: one is a downstream signal of something else going on, and the other has a treatment story that sounds convincing and has not held up in trials. Knowing what each number is measuring — and what acting on it does and does not achieve — is what makes them worth ordering.

What "chronic low-grade inflammation" actually means

Inflammation is not one thing. The acute version is the immune response to an injury or infection: local, intense, purposeful, and over within days. The chronic low-grade version is different in kind, not just in degree — a persistent, systemic, low-amplitude activation with no target to resolve against.

Its main sources are metabolic rather than infective. Visceral adipose tissue is not inert storage; it is secretory, and it releases inflammatory cytokines continuously in proportion to how much of it there is. Insulin resistance both drives and is driven by that signalling. The result is a slow background process that damages endothelium, promotes plaque instability, and contributes to the metabolic and cognitive trajectories that show up decades later. Inflammation as the master disruptor covers that systemic picture.

hs-CRP, and what it is a proxy for

C-reactive protein is made by the liver in response to interleukin-6. The high-sensitivity assay measures the low concentrations relevant to cardiovascular risk, as distinct from the very high values seen in acute illness.

The key point is that CRP is a reporter, not a driver. It reflects IL-6 activity, which reflects whatever is producing the IL-6 — most often visceral fat and insulin resistance, sometimes periodontal disease, sometimes an autoimmune process, sometimes a recent infection. The number tells you the alarm is sounding; it does not tell you where the fire is. hs-CRP independently predicts cardiovascular events even where cholesterol is unremarkable (Ridker et al., NEJM 2008), which is what makes it worth adding to a panel — it identifies risk that a lipid panel alone misses. The deeper hs-CRP article goes further into assay behaviour and pitfalls.

Is inflammation causal, or just a marker?

This is the question that determines whether treating the number is worth anything, and the answer is more interesting than either extreme.

Lowering CRP itself does nothing — genetic variants that alter CRP concentration are not associated with altered cardiovascular risk, which argues strongly that CRP is a passenger. But targeting the inflammatory pathway upstream of it does appear to matter: a randomised trial of an interleukin-1β antibody in patients with prior myocardial infarction and elevated hs-CRP reduced recurrent cardiovascular events without changing lipids at all (Ridker et al., N Engl J Med 2017).

So inflammation is causal; CRP is how you see it. That distinction has a direct practical consequence — the thing to change is the driver, not the reading.

Homocysteine, and what it is really telling you

Homocysteine is an intermediate in methionine metabolism. It has two disposal routes: remethylation back to methionine, which requires folate and B12, and transsulphuration to cysteine, which requires B6. When those routes are under-supplied, homocysteine accumulates.

Because it sits at that junction, homocysteine is as much a readout of B-vitamin status and methylation capacity as it is a vascular risk marker. An elevated value should prompt a look at B12, folate, renal function, thyroid status and medications that interfere with folate metabolism — the methylation article covers that workup.

The homocysteine result that gets left out

Elevated homocysteine predicts events. Lowering it with B vitamins has not reliably prevented them. Large randomised trials and their meta-analyses consistently show that supplementation reduces homocysteine concentration substantially and does not produce the expected reduction in myocardial infarction or cardiovascular death (Clarke et al., Arch Intern Med 2010), a conclusion echoed by systematic review (Martí-Carvajal et al., Cochrane review). Stroke is the partial exception, where some benefit has been reported.

This is the same lesson as CRP, arriving from the other direction. A number that marks a process is not automatically a lever on that process. The honest framing is that homocysteine is useful as a signal — it flags B12 or folate insufficiency, impaired renal clearance, or a methylation problem worth understanding — and that correcting a genuine deficiency is worth doing on its own terms. Treating the number in someone with no deficiency, on the expectation of cardiovascular benefit, is not supported.

What actually drives these numbers up

• Visceral adiposity — the single largest contributor to hs-CRP in most people
• Insulin resistance, which travels with it in both directions (reversal protocol)
• Periodontal disease — a genuinely under-appreciated source of systemic CRP
• Chronic or smouldering infection
• Poor or fragmented sleep, which raises inflammatory signalling within days
• Smoking, and heavy alcohol use
• Autoimmune disease, which changes the interpretation entirely
• B12, folate or B6 insufficiency, and reduced renal clearance, for homocysteine specifically

How to draw them and how to read them

Both markers are easy to misread, in predictable ways.

• Do not draw hs-CRP during or shortly after an illness, injury or hard training block. Any of those will produce a transient elevation that has nothing to do with baseline risk.
• Never act on a single elevated value. Repeat it a couple of weeks later, when well. Confirm before you conclude.
• A value above 10 mg/L is not a risk score. It is a prompt to find the acute cause.
• Read them alongside the rest of the panel, not instead of it. Inflammation and particle burden are separate risk streams that add rather than substitute — see ApoB.
• Track change, not absolutes. Two values six months apart under the same conditions say more than one value ever will.

What actually lowers them, in order of effect

1. Lose visceral fat. The largest single lever on hs-CRP, because it removes the tissue generating the signal.
2. Train, both ways. Resistance training and steady aerobic work each reduce inflammatory markers, and they do it through partly different routes — the aerobic side and a structured full-body strength programme between them cover most of it.
3. Fix sleep, including screening for sleep-disordered breathing if it is unrefreshing.
4. Treat gum disease. Unglamorous, and it moves CRP.
5. Correct genuine B-vitamin deficiency where homocysteine is elevated — for the deficiency, not for the number.
6. Eat a diet built on vegetables, fish and olive oil, low in ultra-processed food. The effect is real and slower than the first item on this list.
7. Stop smoking; moderate alcohol.

Whether any of this belongs in a wider plan, and what else should be measured alongside it, is a physician's judgement after evaluation — the 60-second assessment is the starting point.

Bottom line

hs-CRP and homocysteine are among the most predictive and least-ordered labs in routine practice, and both are markers rather than levers. Inflammation itself is causal in cardiovascular disease — demonstrated by trials that reduced events by targeting the pathway without touching lipids — but CRP is how you observe it, not what you treat. Homocysteine flags a real problem in methylation and B-vitamin status, yet lowering it pharmacologically has not delivered the cardiovascular benefit the association implied. Order both, draw them when you are well, repeat before acting, and put the effort into the drivers underneath: visceral fat, insulin resistance, sleep, and the deficiencies that are actually present.

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