Our April 2003 NewsPath letter to our doctors, about false positive problems

If you use hs-CRP as a cardiac risk screening test, there is enough natural variability of the level day-to-day that one should never decide a risk status based upon just one result. Do at least 2 tests and average the results⁹. It is a stable molecule. Low-risk is less than 1.0 mg/L; intermediate is between 1.0-3.0; and high risk is greater than 3.0⁹. Some consider 0.4 to be the upper limit of "normal"¹⁰.

At LMC, the CRP became a 3 shifts per day, 7 days per week in March 2008. C-reactive protein (CRP) is the first of the "acute phase reactant" proteins to increase during states of acute inflammation. A high or increasing amount of CRP in your blood suggests that you have an acute infection or inflammation (to include inflammed, sterile tumor necrosis or gouty pyoarthrosis [L08-12261]). That patients have high CRP during "attacks" of familial Mediterranian fever (FMF) further assures that elevations do not implicate bacterial infections...they implicate inflammatory reactions that contain polys. If the CRP level in your blood drops, it means that you are getting better and inflammation is being reduced. CRP levels can be elevated in the later stages of pregnancy as well as with use of birth control pills or hormone replacement therapy (i.e., estrogen). Higher levels of CRP have also been observed in the obese.

CRP is exclusively produced in the liver. In this context of rule in or rule out infection, less than 8 mg/L is considered "normal". CRP elevation  is thought to correlate with endothelial dysfunction⁶. The plasma level can double every 8 hours, peaking in 50 hours; upon removal of the stimulus, CRP drops rapidly, having a 5-7 hour half life¹. In the post-shock (trauma, sepsis, etc.) period, the body swings into a catabolic phase and makes acute phase reactants at the expense of such proteins as prealbumin and albumin⁷. A sarcoidosis study normal control group ranged in CRP levels from 3.7-5.9 mg/L¹³.

More recently, related to theories that CRP is a surrogate reflector of the inflammatory component  of the pathogenesis of coronary artery obstructive/stenotic lesions, it has been noted that screening risk of future initial acute myocardial infarction (AMI) is increased in persons with sustained levels of CRP above 2.0 mg/L. This "cardio-CRP" screen (CRP-H...high-sensitivity CRP...hs-CRP, detect down to about 0.1 mg/L) should, like lipid profiles, be done only during periods of good health, free from any recent minor or major illnesses. NOTE: tests at different times in the same non-sick person may vary by about 30%⁴ or more¹⁰; so, as with cholesterol screening, it is best to test your hs-CRP no less than on two, separate healthy occasions. Aspirin, although having anti-inflammatory properties, does not lower hs-CRP¹⁰.

Some (clinical chemist, Nader Rifai, of Harvard) feel that hs-CRP is the single best predictor of future cardiovascular disease risk (those with CRP levels in the upper 25% are 3 times as likely to suffer a heart attack⁶). Sustained elevations are also predictive of increased stroke and peripheral vascular disease risk; and increased levels may predict increased risk of hip fracture in elderly women¹⁴. In healthy adults, 75% of hs-CRP values are < 0.32 mg/L³. The median normal concentration of CRP is 0.8 mg/L, with 90% of apparently healthy individuals having a value less than 3 mg/L and 99% less than 12 mg/L¹. Our lab's "normal" is <0.2 mg/dl (2 mg/L), but there is significant personal day-to-day variation (and note 2nd paragraph, above).

And, different methodology in different labs will likely not relate exactly to published quintile and quartile information⁶...therefore, hs-CRP risk categorization should be viewed somewhat loosely. AHA/CDC 2003 state divides into tertiles & says <1 mg/L is "low risk", 1-3 mg/L is "moderate risk", and >3 mg/L is "high risk"⁹; these are the ranges reported in our lab.

In situations (such as colonic diverticulosis) which have had a pyogenic complication adequately & conservatively managed, one can determine that patient's baseline CRP when perfectly well. Then if suspect symptoms begin, a significantly elevated CRP would suggest prompt re-institution of conservative management again.

CRP level below normal: consider protein calorie malnutrion (PCM).

Situations with CRP Within Normal Range:

• lowest cardio-risk quintile (quintile #1) is < 0.55 mg/L (or 0.1-0.6)...(low risk)

• low-mid (low) cardio-risk (quintile #2) is 0.55-0.99 mg/L (or 0.7-1.1)...(low risk)

• mid-high (moderate) cardio-risk (quintile#3) is 1.0-2.1 (or 1.2-1.9) mg/L...(intermediate risk)

• is recommended as an additional  screening test for CVD in those at intermediate risk (10-20% risk of attack of CVD in the next 10 years [calculate by clicking on "health tools", then "risk assessment"]) to  decide additional diagnostic assessment (imaging...exercise testing) or therapeutic intervention (lipid-lowering, antiplatelet, cardioprotective agents)⁹.

• even such smoldering important but non-acutely threatening active chronic infections as gingivitis, periodontitis, or localized peri-apical bone inflammation can cause elevations into one of the higher quintiles.

• a normal CRP is highly unlikely in the face of a significantly acutely threatening bacterial infection¹.

• in evaluating a patient with "fever and night sweats", the signs and symptoms are almost certainly NOT due to hidden infection when CRP is in normal range (therefore, symptoms & signs could be due to malignant lymphoma).

• may be normal in both temporal arteritis and polymyalgia rheumatica¹¹.

• detecting supervening infection after "clean" surgery: a potentially valuable marker to serially follow after surgery...in search of early detection of postoperative, initially occult (hidden) infection. Clean CRP pattern: begin increase from baseline in 4-6 hours postop & peaks at (usually to intermediate elevations) 48-72 hours, begins to decrease after 72 hours and back near baseline by 5th-7th postop day⁵.

• if serum CRP is not elevated, a failed joint prothesis is unlikely due to sepsis [HERE].

• stable or even progressing, mild sarcoidosis can have CRP less than 8...but very likely above 5...and sarcoidosis unlikely if below 5 mg/L¹³.

Situations with CRP Elevations:

• with cardio-risk screening in mind, values greater than 15 mg/L suggest that the screening effort is confounded by false positive situations (see link above) or recent or acute inflammatory disease beyond what could be expected related to coronary artery disease⁶; post-shock (trauma, sepsis) catabolic phase can highly elevate CRP⁷

• those with troponin-normal cardiac events who have hs-CRP >3 mg/L at discharge have increased risk of major event within the next 3 months and, if >5 mg/L on admission, an increased risk of a major event on out for 6 months⁶

• hi-risk cardio-risk (quintile #4) is 2.0-3.8 mg/L...(high risk if greater than 3.0⁹)

• highest cardio-risk zone (quintile #5), is from 3.9-15 mg/L...(high risk⁹)..."high risk" levels reflect 4-7x normal risk¹⁰.

• even such smoldering important but non-acutely threatening active chronic infections as gingivitis/periodontitis, chronic prostatitis , chronic gastritis, chronic bronchitis, and urinary tract infections can cause elevations into one of the higher quintiles⁹.

• hs-CRP predicts  a future stroke, AMI, or PVD event in males & females, smokers & nonsmokers, when in the highest quartile at a relative risk over "normals" of 2-fold, 3-fold, and 4-fold, respectively.

• usually elevated in temporal arteritis

• hs-CRP joined with TC:HDL-C ratio gives strongest predictive information⁶

• estrogen replacement therapy (HRT or ERT) increases CRP level⁶;  vitamin B6 deficiency will cause CRP "elevations" (probably meaning as to cardio-risk quintiles); elevated blood pressure, elevated BMI (obesity)[calculate your BMI], cigarette smoking, "metabolic syndrome¹²" (see below), diabetes mellitus, low HDL/high triglyceride lipid profile, estrogen/progesterone hormone use, and chronic inflammatory conditions (such as rheumatoid arthritis) all can innately cause elevations of hs-CRP in the absence of inflammation⁹. 

• minor elevations¹: SLE, scleroderma (systemic sclerosis), dermatomyositis, ulcerative colitis, leukemia, GVHD (graft vs. host disease)

• major serum/plasma elevations:¹

• bacterial vs. viral infection: a very high level (>100 mg/L) more likely bacterial than viral, while intermediate levels (10-50 mg/L) can be either viral or bacterial

• postoperative elevations, especially when they can be concretely related to a normal or reasonably low-elevated preoperative CRP baseline test value, tend to be reflective of infection and not just organizing resolution of clean-surgical tissue damage

• monitoring extent & therapeutic response to inflammatory disease:

  • postoperative "dirty case" monitor

  • is the infection responding to the antibiotic (pyelonephritis, pelvic infections, osteomyelitis, meningitis, endocarditis)? CRP not affected by drugs or fever suppressants¹.

• detection and management of intercurrent infections: in diabetics, lupus, ulcerative colitis, Crohn's disease, a major rise above usual baseline CRP level suggests an infection¹

• An analysis of 3597 women with metabolic syndrome showed that those with the highest CRP levels (greater than 3.0 mg/L) were 2.1 times more likely to have a cardiovascular event within 8 years than those with the lowest CRP levels (less than 1.0 mg/L)¹².

References:

1. Hunter Area Pathology Services...click on "information sheets" & pick the test, Nov. 2001 [site]
2. Specialty Labs manual, Jan 2001.
3. Mayo Medical Lab manual, Jan 2001.
4. Serum Markers of Risk for Coronary Artery Disease, Plaut D, Advance for ...., 10(11):41-44, Nov. 2001.
5. Interpretation of Diagnostic Tests, Wallach, 2000, 7th Ed., pages 843-847.
6. Rafai & Ridker, Clinical Chemistry, 47(3):403-411, 2001
7. Proteins Used in Nutritional Assessment, Spiekerman AM, Clinics In Lab. Medicine 13(2):353-369, June 1993..
8. NEJMed, November 2002 
9. Pearson TA, et. al., Markers of Inflammation and Cardiovascular Disease...[AHA/CDC Scientific Statement], Circulation 107:499-511, 28 January 2003.
10. Superko HR, Before The Heart Attacks, 334 pages, 2003.
11. Epperly TD, et al. (2000) Polymyalgia rheumatica and temporal arteritis. American Family Physician, 62(4): 789–796.
12. Julie E. Buring, M.D.; Nancy R. Cook, M.D.; and Nader Rifai, M.D., Circulation: Journal of the American Heart Association, 2003.
13. Eur Respir J 2003; 21:407-413 Copyright ©ERS Journals Ltd 2003.
14. letter to editor, page 1353, JAMA 296(11):1353-54, 20 Sept. 2006.
    

(posted Jan. 2001; latest addition 13 June 2009)