C-Reactive Protein

C-reactive protein (CRP) is an acute phase protein that is produced in the liver. It is a major acute phase protein in the dog but also works well as a marker of inflammation in rabbits (Cray et al 2015). Concentrations in healthy dogs are quite low but marked increases (over 50 fold) occur rapidly in response to acute inflammation. Concentrations also decrease rapidly with resolution of inflammation, thus monitoring serial changes over time can be helpful with documenting resolution or worsening of inflammation in a given animal. Acute phase proteins are considered part of the innate immune response and have antibacterial effects. Indeed, CRP was named by its ability to bind to techoic acid (also known as C-polysaccharide) in the bacterial cell wall of Pneumococcus. Through binding to the cell wall, CRP activates the classical pathway of complement (C1q), leading to bacterial opsonization, which promotes clearance by phagocytes. It also binds to other ligands, including phosphatidylcholine and membrane lipids and DNA in damaged cells.


C-reactive protein consists of 5 subunits of around 20 kDa each, forming a pentamer. It is produced in the liver in response to inflammatory cytokines, such as interleukin-1, interleukin-6 and tumor necrosis factor-α. It migrates in the gamma region of the electrophoretogram, but concentrations are too low to result in changes in the electrophoretogram, even when marked increases occur.


There are various methods to measure CRP, all of which are immunologic-based, including ELISA, immunoturbidometric assays, latex agglutination and a time-resolved immunofluorometric assay. The assay at Cornell University is an automated particle-based immunoturbidometric assay (Gentian canine CRP assay), which uses canine-specific antibodies against CRP. This has been previously validated (Hillström et al 2014). In house testing revealed adequate performance of the assay. Point of care tests are also available (Jasensky et al 2015) but vary in their performance (and should not be used interchangeably).

Reaction type



  • Method: The reagent contains particles with bound anti-canine CRP antibodies. The beads bind CRP in the patient’s sample, forming a precipitate, which alters the turbidity of the sample. The change in turbidity is measured spectrophotemetrically, after calibration with a standard with known CRP concentrations (in this case, presumably a canine standard).
  • Limits of quantification: 5.3 to 300 mg/L (per package insert). Results above the upper limit are diluted to bring results within the quantification limits. We serially diluted a sample with high CRP concentration and found linearity between 0.3 and 209 mg/L
  • Lower limit of detection (mean ± 3 SD of blank): 2.5 mg/L.
  • Precision: With our Cobas C501, the intra-assay coefficient of variation is 1-1.2 % (sample concentrations of 26 and 183 mg/L) with inter-assay coefficient of variation ranging from 1.3 to 6.2% (the latter was from a sample with low results of 13 mg/L).

Units of measurement

CRP concentration is measured in mg/L (SI units) and μg/mL (conventional units). We use SI units.

Conversion: 1 mg/L = 1 μg/mL

Sample considerations

Sample type

Serum, heparinized and EDTA plasma (per package insert). C-reactive protein has been measured in saliva, using a time-resolved immunofluorometric assay (Parra et al 2005), in joint fluid, using ELISA, (Boal and Miguel Carreira 2015) and in spinal fluid (Anderson et al 2015).


Our in-house data shows that CRP can be measured in serum and heparinized plasma with expected results, although we did not do direct comparisons between these two sample types in the same animal.


CRP was stable for storage at 14 days refrigerated or at room temperature. Sample could be frozen and thawed for 4 cycles (Hillström et al 2014). In an additional study with a higher sensitivity assay (this was accomplished by adding 10x more sample than normally used), CRP was stable in frozen samples stored in a dedicated freezer at -80°C for 3 months (Hillström et al 2015).


  • Lipemia, hemolysis, and icterus: Per the validation manuscript, addition of hemoglobin (from osmotically lysed red blood cells) and lipid (intralipid) did not affect concentrations(Hillström et al 2014). The effect of icterus is unknown, but is expected to be minimal.
  • Drugs: None reported. Unlike haptoglobin, corticosteroids (single dose of 1.1 mg/kg methylprednisolone SC and 1 to 2.2. mg/kg prednisone orally for 3 weeks to 7 days, respectively, did not increase CRP concentrations in dogs (Martinez-Subiela et al 2004).

Test interpretation

Hillström et al. (2014) established reference intervals from 40 dogs (ranging from 0.5-11 years). All dogs except one had concentrations < 6.8 mg/L. One dog had a high value of 16 mg/L, which was decreased after frozen storage and could have been a falsely high result (or falsely decreased with storage). Concentrations of <20 mg/L were found in clinically healthy Beagle dogs with an ELISA (Kuribayashi et al 2003) and Miniature Schnauzers (Wong et al 2011). Higher reference intervals have been reported with a point of care kit ( Torrente et al 2015). We add a comment to our results stating that concentrations in healthy dogs are usually < 10 mg/L. Due to a low index of variability (0.78), critical differences may be more relevant in interpreting results than a reference interval (Carney et al 2011). A one-sided critical difference of 270% (the degree to which a change in CRP is considered clinically relevant in one direction, i.e. up or down) was established in one study, in which 11 dogs of mixed breed were serially tested over 12 weeks (Carney et al 2011).  In another study of 11 German-Braco dogs, in which blood samples were collected weekly for 5 weeks, an absolute critical difference of 4.9 mg/L was obtained with an ELISA assay (Martinez-Subiela et al 2003). However, this critical difference approach requires obtaining baseline CRP concentrations in a healthy dog that has reached adulthood, something that is rarely done in veterinary medicine.

Increased CRP concentration

  • Physiologic: There are no sex- or age-related differences in healthy beagle dogs (Kuribayashi et al 2013).
    • Pregnancy: High concentrations of CRP can occur in female dogs during pregnancy, with results peaking between 70 and 90 mg/L, within 1 -1.5 months after ovulation (Kuribayashi et al 2013). Healthy dogs maintained in a normal environment had higher CRP concentrations (8.4 ± 4.6 mg/L) than dogs kept in a research “clean” environment (0.5 ± 0.2 mg/L).
    • Exercise: Exercise is associated with an increase in CRP concentrations (Yazwinski et al 2013Fergestad et al 2016).
  • Pathophysiologic: High CRP concentrations are generally considered a sensitive marker of inflammation in various conditions.
    • Inflammation: CRP concentrations will increase with local or systemic inflammation but increases are not specific as to cause (Christensen et al 2014, Torrente et al 2015, Behling-Kelly et al 2022). Significantly higher mean concentrations were seen in dogs with systemic inflammatory response syndrome or sepsis (165 ± 82 mg/L) than localized inflammation (108 ± 70 mg/L) in one prospective study of 116 dogs (Torrente et al 2015), although there was overlap between groups. In addition, some dogs in the negative control group (blood donors) had high concentrations (Torrente et al 2015). In the same study, CRP concentrations had good discriminatory ability (based on receiver operator characteristic curves, with area under the curve of 0.932 to 0.996) between dogs with inflammation (localized and systemic) and blood donor controls but weaker ability to discriminate between localized and systemic inflammation  (area under curve of 0.704) and survivors and non-survivors (area under curve of 0.554). Plasma iron (decreased) and fibrinogen (increased) concentrations (Clauss or clotting method) showed similar discriminatory ability as CRP for inflammation.
    • Immune-mediated hemolytic anemia: Dogs with IMHA have high values of CRP (143 ± 89 mg/L) on day 1 of admission, which normalizes with treatment, illustrating that sequential CRP measurement can show resolution of inflammation. Criteria for inclusion were dogs with anemia (≤30% HCT) with at least one of persistent agglutination (slide test), positive Coombs test, or moderate to marked spherocytosis (Mitchell et al 2009).
    • Cancer: High CRP concentrations are reported in dogs with various forms of cancer, particularly that of hematopoietic origin (lymphoma, leukemia, including chronic lymphocytic leukemia, acute lymphoid and myeloid leukemia). The increases could be due to tumor-associated inflammation or cytokine production (either produced by the tumor or host cells as an anti-tumor response) (Tecles et al 2005, Mischke et al 2007Nielsen et al 2007).
    • Other conditions: CRP will increase with various types of insults, including trauma (such as post-surgery), inflammation, due to infectious or non-infectious causes, and congestive heart failure (Christensen et al 2015, Reimann et al 2016). Measurement of CRP is not useful for identifying the specific disease or cancer relapse in dogs with lymphoma (Merlo et al 2007).


Scroll to Top