What an Inflammation Panel Measures, and How to Read One

Oak
inflammation biomarkers crp proteomics blood-testing
Five glowing columns of red and amber fluid at different heights inside a dark laboratory instrument.

An inflammation panel is a small set of blood analytes that rise when the innate immune system is activated: most commonly high-sensitivity C-reactive protein (hs-CRP), erythrocyte sedimentation rate (ESR), fibrinogen, ferritin, and interleukin-6 (IL-6), usually alongside a CBC with differential. None of these markers tells you where inflammation is or what is causing it. They are non-specific amplitude readings. A single elevated hs-CRP after a head cold, a hard interval session, or a poor night of sleep means almost nothing. A stable, repeatedly measured hs-CRP of 4 mg/L in an otherwise asymptomatic person is a different kind of signal, and one worth bringing to a physician rather than interpreting yourself.

The useful framing for a technically fluent person: these are low-dimensional summary statistics of a high-dimensional process, with large within-person variance and strong confounding by body composition. Treat them as time series, not as point readings.

The markers a standard panel contains

hs-CRP. An acute-phase pentraxin made in the liver, driven largely by IL-6. Plasma half-life is about 19 hours, so the level tracks IL-6 production over roughly the preceding day or two. It spans five orders of magnitude, from ~0.1 mg/L to >200 mg/L in sepsis. “High sensitivity” refers only to assay precision at the low end; it is the same analyte. Standard cardiovascular risk stratification (AHA/CDC) uses <1, 1–3, and >3 mg/L, and explicitly says to discard any value above 10 mg/L as likely acute and repeat later.

ESR. Not an analyte at all. It is the sedimentation rate of red cells in a Westergren tube over one hour, driven mostly by fibrinogen and immunoglobulins altering rouleaux formation. It responds slowly (days to weeks), which makes it a useful low-pass complement to CRP, and it is confounded by anemia, red cell morphology, age, sex, and pregnancy.

Fibrinogen. Acute-phase clotting factor, typical adult range 200–400 mg/dL. Slower and lower-dynamic-range than CRP.

Ferritin. Iron storage protein and an acute-phase reactant, which is why it is unreliable as an iron status marker in anyone with active inflammation. Interpreting ferritin without a concurrent CRP is an error.

IL-6. The upstream cytokine driving most of the hepatic acute-phase response. Circulating concentrations in healthy people sit in the low pg/mL range, which puts it at the edge of conventional ELISA sensitivity and makes it extremely sensitive to sample handling.

CBC-derived ratios (neutrophil-to-lymphocyte, platelet-to-lymphocyte) come free with any differential and have been used as inflammatory readouts in panel studies of environmental exposure, including repeated-measures designs in children exposed to polycyclic aromatic hydrocarbons and phthalates.1

One measurement is mostly noise

Within-person biological variation in hs-CRP is large, comparable to or greater than the between-person variation that defines the risk tertiles. This is the single most important fact about the panel and the one consumer lab pages omit. The operational consequences:

  • Take at least two hs-CRP measurements, at least two weeks apart, under matched conditions, and average them. Three is better.
  • Drop any value >10 mg/L and re-draw after four weeks.
  • Exclude draws within ~2 weeks of any infection, vaccination, dental work, or injury, and within 48–72 hours of unaccustomed eccentric exercise.
  • Standardize time of day and fasting state. Log them either way so you can model them later.

If you are building a personal time series, treat each draw as value ~ intercept + season + acute_event + noise and do not react to single points. A useful discipline: before you look at a new result, write down what change would be large enough to act on. For hs-CRP, a doubling that persists across two consecutive draws is a reasonable threshold for a conversation with a clinician. Anything smaller is inside the noise.

Preanalytical handling decides whether cytokine numbers mean anything

CRP, ESR, fibrinogen, and ferritin are forgiving. Cytokines are not. If your panel includes IL-6, TNF-α, IL-1β, or IL-10, the number you get is partly a measurement of how the tube was handled. A cohort study of pro- and anti-inflammatory biomarker stability found marker-specific sensitivity to processing delay and freeze-thaw, with some analytes tolerating room-temperature holds and others degrading or artifactually rising.2 Ex vivo leukocyte secretion during a delayed spin can create IL-6 that was never in your bloodstream.

Practical rules we follow:

  • EDTA plasma over serum for cytokines. Serum requires a clotting interval during which platelets and leukocytes release proteins.
  • Centrifuge within 30 minutes of draw, 1,500–2,000 × g for 10–15 minutes at 4 °C.
  • Aliquot into single-use volumes (200 µL) and freeze at −80 °C. Every freeze-thaw cycle costs you.
  • Keep one aliquot untouched as a bridging sample for future assays.

Ask any vendor for time-to-spin and storage temperature. If they cannot answer, the cytokine columns in your report are decorative.

What sets your baseline

Before you attribute a CRP of 3.5 mg/L to diet, account for the dominant covariates. In a large analysis of multiple systemic inflammation markers, adiposity was the strongest and most consistent correlate across CRP, IL-6, and soluble TNF receptors, with additional contributions from age, sex, and smoking.3 Adipose tissue is an IL-6 source. Any personal experiment that changes body composition will move CRP, and attributing that movement to a specific food or supplement is a confound, not a finding.

Going past five markers

The standard panel is narrow because the assays are old and cheap. Multiplex affinity proteomics (Olink PEA, SomaScan aptamers, NULISAseq) measure hundreds to thousands of proteins from ~50–100 µL of plasma, including the cytokines, chemokines, and receptors that CRP only summarizes. A NULISAseq inflammation panel applied to Alzheimer’s cohorts resolved individual inflammatory proteins against neurodegeneration markers in the same sample.4 In Parkinson’s disease, blood inflammatory markers measured at baseline associated with long-term clinical severity, which is the kind of longitudinal association a single CRP cannot support.5 And a four-biomarker blood signature has been shown to discriminate systemic inflammation from viral infection versus other causes, which is exactly the specificity CRP lacks.6

These platforms return relative units, not concentrations. Olink gives NPX, a log2-scale normalized value: a difference of 1 NPX is a doubling. Batch effects dominate cross-run comparison, so plan bridging samples (a fixed set of aliquots run in every batch) before you send anything. In R, OlinkAnalyze handles NPX import, LOD filtering, and bridge normalization. Filter proteins below LOD in more than ~50% of your samples rather than imputing them, and correct for multiple testing across the panel. With 384 proteins and no correction, you will find significance in noise every time.

Diet, without overclaiming

Dietary pattern is associated with low-grade inflammation markers, but the effect sizes are modest and the literature is mostly observational. A systematic review of dietary pattern analysis and inflammation biomarkers found consistent associations for patterns characterized by vegetables, fruit, whole grains, and fish versus patterns high in refined grains and processed meat, with CRP and IL-6 as the most frequently reported outcomes.7 Analyses of individual food groups suggest the associations depend on context, including meal composition.8 Higher dietary inflammatory index scores have been associated with greater symptom severity in depressed individuals with appetite loss.9 Randomized evidence is thinner: trials such as FEAST, comparing an anti-inflammatory diet with a low-fat diet in knee osteoarthritis, are still reporting.10

The defensible summary: no single food reliably moves your CRP, and the “worst foods for inflammation” listicle genre is not supported by measurement-grade evidence. If you want to test a dietary change on yourself, run it as a design with a baseline of three pre-intervention draws, an intervention of at least eight weeks, and three post draws, with body weight logged throughout as a covariate.

The protocol we’d run

  1. Baseline: CBC with differential, hs-CRP, ESR, fibrinogen, ferritin plus iron studies, and a broad affinity proteomics panel on EDTA plasma processed within 30 minutes.
  2. Repeat hs-CRP and CBC at weeks 2 and 4 to establish your personal within-subject variation. Everything after that is compared against your own distribution, not a population reference interval.
  3. Bank duplicate −80 °C aliquots at every draw.
  4. Keep a structured log: draw time, fasting hours, last hard training session, sleep the prior night, any illness or vaccination in the past 30 days, current weight.
  5. Persistently elevated hs-CRP (>3 mg/L across repeat draws), any value >10 mg/L that does not resolve, or an elevated ESR with symptoms belongs with a physician. Marked or persistent elevations can accompany infection, autoimmune disease, and malignancy, and sorting that out requires clinical examination and imaging, not a spreadsheet.

Questions people also ask

What are the five markers of inflammation? hs-CRP, ESR, fibrinogen, ferritin, and IL-6 are the five most panels mean. Procalcitonin is sometimes added in acute care settings, and CBC-derived ratios like neutrophil-to-lymphocyte come with any differential.

Is inflammation in a blood test serious? A single mildly elevated marker usually is not, and often reflects a recent infection, hard exercise, or higher body fat. A high value that persists across repeat draws, or a very high value (hs-CRP >10 mg/L) that does not resolve, needs clinical evaluation, because the panel cannot tell you the source.

What diseases have high inflammatory markers? Bacterial and viral infection, rheumatoid arthritis and other autoimmune conditions, inflammatory bowel disease, cardiovascular disease, and many cancers. A four-marker blood signature can separate viral from non-viral systemic inflammation, which the standard panel cannot.6 Neurodegenerative conditions also show inflammatory protein associations in blood.45

What are the 5 classic signs of inflammation? Redness, heat, swelling, pain, and loss of function, the classical descriptions of local acute inflammation. They describe visible tissue-level inflammation and have little to do with the low-grade systemic inflammation an hs-CRP of 2.5 mg/L represents.

What are the 5 worst foods for inflammation? There is no measurement-grade list. The evidence supports whole dietary patterns over individual foods, with patterns high in refined grains and processed meat associated with higher CRP and IL-6, and vegetable, fruit, whole-grain, and fish patterns associated with lower levels.78

What does it mean if my bloodwork shows inflammation? It means an acute-phase response was active in the day or two before the draw. It does not identify a location or a cause. Repeat the measurement under matched conditions before concluding anything, and bring persistent elevations to a clinician.

Oak builds longitudinal molecular profiles of individuals: whole-genome sequencing, RNA sequencing, proteomics, blood biomarkers, and continuous glucose data, integrated into one model of you. Build your profile.

Footnotes

  1. Lei Zhao, Miao Liu, Linlin Liu, et al. The association of co-exposure to polycyclic aromatic hydrocarbon and phthalates with blood cell-based inflammatory biomarkers in children: A panel study. Environmental Pollution, 2022. https://doi.org/10.1016/j.envpol.2022.119479 ↩

  2. C. Graham, R. Chooniedass, W. P. Stefura, et al. Stability of pro- and anti-inflammatory immune biomarkers for human cohort studies. Journal of Translational Medicine, 2017. https://doi.org/10.1186/s12967-017-1154-3 ↩

  3. Sandi L. Navarro, Elizabeth D. Kantor, Xiaoling Song, et al. Factors Associated with Multiple Biomarkers of Systemic Inflammation. Cancer Epidemiology, Biomarkers & Prevention, 2016. https://doi.org/10.1158/1055-9965.epi-15-0956 ↩

  4. Guglielmo Di Molfetta, Ilaria Pola, Kubra Tan, et al. Inflammation biomarkers and Alzheimer’s disease: A pilot study using NULISAseq. Alzheimer’s & Dementia: Diagnosis, Assessment & Disease Monitoring, 2025. https://doi.org/10.1002/dad2.70079 ↩ ↩2

  5. Dagmar H. Hepp, Thecla A. van Wageningen, Kirsten L. Kuiper, et al. Inflammatory Blood Biomarkers Are Associated with Long-Term Clinical Disease Severity in Parkinson’s Disease. International Journal of Molecular Sciences, 2023. https://doi.org/10.3390/ijms241914915 ↩ ↩2

  6. D. L. Sampson, B. A. Fox, T. D. Yager, et al. A Four-Biomarker Blood Signature Discriminates Systemic Inflammation Due to Viral Infection Versus Other Etiologies. Scientific Reports, 2017. https://doi.org/10.1038/s41598-017-02325-8 ↩ ↩2

  7. Janett Barbaresko, Manja Koch, Matthias B Schulze, et al. Dietary pattern analysis and biomarkers of low-grade inflammation: a systematic literature review. Nutrition Reviews, 2013. https://doi.org/10.1111/nure.12035 ↩ ↩2

  8. Carolina Schwedhelm, Lukas Schwingshackl, George O. Agogo, et al. Associations of food groups and cardiometabolic and inflammatory biomarkers: does the meal matter?. British Journal of Nutrition, 2019. https://doi.org/10.1017/s000711451900151x ↩ ↩2

  9. Kaiping Burrows, Jennifer L. Stewart, Chase Antonacci, et al. Association of poorer dietary quality and higher dietary inflammation with greater symptom severity in depressed individuals with appetite loss. Journal of Affective Disorders, 2020. https://doi.org/10.1016/j.jad.2019.11.160 ↩

  10. Lynette Law, Joshua L Heerey, Brooke L Devlin, et al. Effectiveness of an anti-inflammatory diet versus low-fat diet for knee osteoarthritis: the FEAST randomised controlled trial protocol. BMJ Open, 2024. https://doi.org/10.1136/bmjopen-2023-079374 ↩