Clinical Chemistry Link to Randox Laboratories Web Site
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Clinical Chemistry 49: 1258-1271, 2003; 10.1373/49.8.1258
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit an electronic Letter to
the Editor about this paper
Right arrow View responses
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (54)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ledue, T. B.
Right arrow Articles by Rifai, N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ledue, T. B.
Right arrow Articles by Rifai, N.
Related Collections
Right arrow Lipids, Lipoproteins, and Cardiovascular Risk Factors
(Clinical Chemistry. 2003;49:1258-1271.)
© 2003 American Association for Clinical Chemistry, Inc.

Review

Preanalytic and Analytic Sources of Variations in C-reactive Protein Measurement: Implications for Cardiovascular Disease Risk Assessment

Thomas B. Ledue1,a and Nader Rifai2

1 Foundation for Blood Research, Scarborough, ME 04070-0190.

2 Children’s Hospital and Harvard Medical School, Boston, MA 02115.

aAddress correspondence to this author at: Foundation for Blood Research, 69 US Route One, Scarborough, ME 04070-0190. Fax 207-883-1377; e-mail tledue{at}fbr.org.


   Abstract
Top
 Abstract
Introduction
 Structure
 Genetics
 Physiologic Function
 Clinical Significance
 Sources of Variability in...
 Conclusions
 References
 
Background: C-reactive protein (CRP) is a widely recognized indicator of inflammation and is known to play an important role in atherogenesis. Recent prospective studies have demonstrated that increased CRP concentrations within the reference interval are a strong predictor of myocardial infarction, stroke, sudden cardiac death, and peripheral vascular disease in apparently healthy adults. On the basis of available evidence, the American Heart Association and the CDC have issued guidelines for the utility of CRP in the primary prevention of coronary heart disease and in patients with stable coronary disease or acute coronary syndromes. Nevertheless, there remains considerable work to optimize the utility of this marker for risk assessment.

Issues: Most traditional CRP tests designed to monitor acute and chronic inflammation have inadequate sensitivity for risk stratification of coronary disease. Thus, manufacturers have had to develop tests with higher sensitivity. Because an individual’s CRP concentration will be interpreted according to fixed cut-points, issues related to the preanalytic and analytic components of CRP measurement must be considered and standardized where possible to avoid potential misclassification of cardiovascular risk.

Conclusions: Efforts to define performance criteria for high-sensitivity CRP applications coupled with growing awareness of the physiologic aspects of CRP most likely will lead to refinements in standardization, improved performance in quality-assessment schemes, and enhanced risk prediction.


   Introduction
Top
 Abstract
 Introduction
Structure
 Genetics
 Physiologic Function
 Clinical Significance
 Sources of Variability in...
 Conclusions
 References
 
In 1930, Tillet and Francis (1) observed a substance in the serum of individuals with Pneumococcus infections that formed a precipitate when mixed with the C-polysaccharide coat of Streptococcus pneumoniae. They noted that this "C-reactive" activity was absent from the sera of healthy individuals. MacLeod and Avery (2) subsequently characterized this substance as a protein and introduced the term "acute phase" to describe the serum of patients with various acute infections. Shortly thereafter, Lofstrom (3) demonstrated the presence of the acute-phase response (APR)1 in both acute and chronic inflammatory conditions; consequently, C-reactive protein (CRP) became recognized as a nonspecific acute-phase protein. This protein has been highly conserved during evolution; it has the same functional and structural homology as a protein that is detected in high concentrations in the hemolymph of the horseshoe crab (Limulus polyphemus), a "living fossil" (4).


   Structure
Top
 Abstract
 Introduction
 Structure
Genetics
 Physiologic Function
 Clinical Significance
 Sources of Variability in...
 Conclusions
 References
 
CRP belongs to a family of pentameric proteins known as pentraxins. It is composed of five identical, noncovalently bonded subunits, and each subunit consists of 206 amino acid residues with a calculated molecular mass of 23 017 kDa; therefore, the total molecular mass of CRP is ~118 000 kDa. This arrangement is very similar to that of another acute-phase protein known as serum amyloid P component (5). The structure of CRP contains a crystal contact where the calcium-binding loop from one protomer coordinates into the calcium site of a second protomer to form the pentameric structure (Fig. 1 ). This configuration allows for the binding of the ligand phosphocholine and provides information concerning conformational changes related to calcium binding.



View larger version (124K):
[in this window]
[in a new window]
  		Figure 1. Pentameric structure of human CRP.

For a further description of the model, see Shrive et al. (131). Image copyright: Keele University, UK. Published with permission by the Nature Publishing Group (http://www.nature.com/).


   Genetics
Top
 Abstract
 Introduction
 Structure
 Genetics
Physiologic Function
 Clinical Significance
 Sources of Variability in...
 Conclusions
 References
 
The gene for CRP is located on the proximal long arm of chromosome 1, as are the inflammation-related genes for serum amyloid P component and Fc receptors. The gene for CRP, found at 7.7 kb on the chromosome and 2.5 kb long, codes for the 206 amino acid residues and comprises two exons separated by a single intron. The first two amino acids are encoded by the first exon, and the remaining amino acids are encoded by the second exon (6). A significant association between CRP genotypes and CRP concentrations has been documented (7). Furthermore, data from the Family Heart Study have revealed that heritability estimates for CRP are 35–40% (8); no known deficiency states have been described for CRP.


   Physiologic Function
Top
 Abstract
 Introduction
 Structure
 Genetics
 Physiologic Function
Clinical Significance
 Sources of Variability in...
 Conclusions
 References
 
The precise in vivo role of CRP is not completely understood, but its properties are consistent with a fundamental role as a nonspecific defense mechanism. In response to tissue injury or infection, CRP synthesis occurs in hepatocytes whose activity is stimulated by cytokines, especially interleukin (IL)-6, IL-1ß, and tumor necrosis factor-{alpha} (9). In the presence of calcium ions, CRP binds to polysaccharides of many bacteria, fungi, and certain parasites. In addition, CRP also binds to phosphorylcholine, phosphatidylcholines, and nucleic acids, and it demonstrates a non-calcium-dependent binding to cationic molecules such as protamine, heparin, and histones (10). More recently, CRP has been shown to bind to various lipid structures, such as liposomes and lipoproteins, which on aggregation are incorporated into LDL and VLDL (11). Once bound, CRP is a powerful activator of the classic complement system and can promote opsonization and phagocytosis of foreign substances. It is one of the most consistently increased and fastest reacting acute-phase proteins (biological half-life of 19 h), suggesting that it is part of the innate immune response (12). Concentrations may increase 1000-fold or more within 24–48 h of tissue injury.


   Clinical Significance
Top
 Abstract
 Introduction
 Structure
 Genetics
 Physiologic Function
 Clinical Significance
Sources of Variability in...
 Conclusions
 References
 
Owing to the speed and magnitude of its response, CRP has been historically used to detect and predict the outcome of various infectious, inflammatory, and necrotic processes and to assess the efficacy of treatment for those processes. Mild inflammation and viral infections generally cause CRP concentrations to increase to ~10 to 50 mg/L, whereas active inflammation and bacterial infections generally cause concentrations between 50 and 200 mg/L (13). Concentrations >200 mg/L are seen in more severe infections and in trauma. As a sensitive but nonspecific marker of inflammation, CRP concentrations should always be interpreted in the context of the patient’s clinical history, preferably with review of previous results.

Recent evidence has clearly demonstrated that increases in CRP concentrations within the reference interval are associated with future coronary events in apparently healthy men and women (14)(15)(16)(17)(18)(19). Those with baseline CRP concentrations in the highest quartile are at two to four times the risk of future myocardial infarction (MI), ischemic stroke, peripheral arterial disease, and sudden cardiac death compared with those with CRP in the lowest quartile. Baseline concentrations of other inflammatory markers, such as serum amyloid A (15), IL-6(19), soluble intercellular adhesion molecule-1 (20), and P-selectin (21), have shown similar association with future coronary events, thus further reflecting the current understanding of vascular biology of atherogenesis. Over the last decade, the contribution of chronic inflammation to the initiation and progression of atherosclerosis has become more understood and appreciated. Compared with other novel and traditional markers of coronary heart disease (CHD), CRP was shown to be the strongest predictor of future coronary events, and when combined with total cholesterol, HDL-cholesterol, and LDL-cholesterol, its ability to predict risk was improved further (14)(15)(16)(22)(23)(24). It is important to note that high-sensitivity (hs) methods are needed for the measurement of CRP for the purpose of assessing risk of cardiovascular disease in apparently healthy individuals. A recent analysis from the Women’s Health Study not only demonstrated that CRP is superior to LDL-cholesterol in predicting future coronary events but also showed that women with high CRP and low LDL-cholesterol are at a higher risk of coronary events than those with high LDL-cholesterol but low CRP (25). In addition, CRP was found to add to the ability to predict risk at any LDL-cholesterol concentration or Framingham Risk Score, indicating that this marker identifies a group of individuals at increased risk who are currently missed under traditional measures (25). Data from the same cohort also demonstrated that CRP adds clinically important prognostic information to the metabolic syndrome; women with metabolic syndrome and increased CRP are at twice the risk of coronary events than those with metabolic syndrome and low CRP (26).

Potential intervention strategies to reduce risk of future coronary events in individuals with increased CRP concentration by use of aspirin (14)(27)(28) or statin (29)(30) have been shown. In addition, CRP has been shown to have prognostic utility in patients with acute coronary syndromes (31), even in the absence of myocardial necrosis, suggesting that CRP may reflect plaque vulnerability and its likelihood to rupture (32)(33). Other inflammatory markers, such as IL-6 and serum amyloid A, have shown similar utility (29)(34)(35)(36). Although its exact role in atherogenesis is not known, current evidence suggests that CRP may be an actual culprit and not simply an innocent surrogate marker for systemic or vascular inflammation. CRP was shown to activate complement, up-regulate the production of adhesion molecules, increase LDL uptake into macrophages, stimulate nitric oxide production and endothelial nitric oxide synthase expression, and increase plasminogen activator inhibitor-1 expression and activity (37)(38)(39)(40)(41).

On the basis of these findings, the CDC and the American Heart Association (AHA) issued guidelines for the utility of this marker in the primary prevention setting and in patients with stable coronary disease or acute coronary syndromes (42). The guidelines also included specific recommendations that pertain to the laboratory aspect of CRP and defined cut-points for clinical interpretation; CRP concentrations <1 mg/L are considered low, 1–3 mg/L average, and >3 mg/L high relative risk.

For any analyte to be measured correctly, a better understanding of the preanalytic and analytic variability is required. Discussed below are the various sources of variability in the measurement of CRP, some of which were considered by the CDC/AHA expert panel in its deliberation (Table 1 ).


View this table:
[in this window]
[in a new window]
  		Table 1. Variables known to affect CRP results.


   Sources of Variability in CRP Measurement
Top
 Abstract
 Introduction
 Structure
 Genetics
 Physiologic Function
 Clinical Significance
 Sources of Variability in...
Conclusions
 References
 
preanalytic variation
Physiologic considerations

Race and ethnicity.
The majority of traditional CRP immunoassays are unable to reliably measure CRP <5 mg/L. Historically, laboratories have reported these values as "less than" the assay’s lower detection limit; however, with the wide availability of hs-CRP methods, the determination of population distributions for CRP is now feasible. Using a particle-enhanced nephelometric assay for hs-CRP, Ledue et al. (43) found that the distribution of CRP concentrations in both genders was nongaussian when evaluated for skewness and kurtosis (Fig. 2 ), consistent with findings elsewhere (25)(44). Recent data from several American and European studies have clearly demonstrated the comparable distribution of CRP concentrations among women not receiving hormone replacement therapy and men (Table 2 and see below) (45)(46)(47). The 50th percentile of CRP measured in the various populations was ~1.5 mg/L for both genders. Furthermore, data from the National Health and Nutrition Examination Survey III showed no significant difference in the distribution of CRP concentration among white, African-American, and Mexican-American men (Table 3 ) (45). Moreover, a comparable CRP distribution was seen in Japanese men (48). Japanese women, however, seem to have slightly lower CRP concentrations. Furthermore, the geometric mean for CRP in Indian Asians was reported to be 17% higher than in European whites (49), a difference that was no longer significant after the adjustment for central obesity and insulin resistance (50). The clinical implication of these findings is that no gender- or ethnic-specific cut-points for CRP are indicated.



View larger version (20K):
[in this window]
[in a new window]
  		Figure 2. Distribution of CRP concentration in 252 apparently healthy adults.

Derived from data reported in Ledue et al. (43).


View this table:
[in this window]
[in a new window]
  		Table 2. Population distributions of CRP (mg/L).1


View this table:
[in this window]
[in a new window]
  		Table 3. Distributions of CRP (mg/L) among men.1

Age and sex.
Most studies have reported no relationship between age (range, 20–70 years) and serum CRP concentrations (43)(44)(51) (Fig. 3 ). However, at least two studies reported a slight increase of CRP concentrations with age (52)(53). The authors could not exclude the possibility that such an increase might be attributable to the increased incidence of obesity that is associated with aging (52) or to subclinical inflammation and seasonal collection issues (53). In the largest study done to date, which included 15 770 women, only a slight change in CRP concentration with age was seen: median CRP concentrations for individuals 45–54, 55–64, 65–74, and >=75 years of age were 1.31, 1.89, 1.99, and 1.52 mg/L, respectively (46).



View larger version (28K):
[in this window]
[in a new window]
  		Figure 3. Distribution of CRP results according to age in 252 apparently healthy adults.

Log CRP = 0.0014 x age + 0.0035 (r = 0.078; P = 0.71). Derived from data reported in Ledue et al. (43).

Seasonal variation.
At present, there are limited data on CRP concentrations and seasonal cycles. In a group of 24 elderly individuals (age >=75 years) whose blood was collected monthly for 1 year, a change of 3.7 mg/L was observed between winter and summer; no evidence for infection was found to explain the difference (54). In contrast, no consistent pattern of change in CRP concentrations was reported from SEASON, a study specifically designed to examine seasonal changes in cardiovascular risk biomarkers (55).

Within- and between-subject variation.
In one study, the within-subject CV for 19 healthy adults studied over 20 weeks was 63% compared with a between-subject CV of 76% (56). The authors of another study reported a within-subject CV of 42% and a between-subject CV of 92% (57). Such variation has led some to question whether it is possible to reliably predict CHD risk using smaller groups such as tertiles, quartiles, or quintiles (58)(59)(60)(61). In contrast, data from one study summarizing the biological variability for CRP indicated that the rather large intraindividual CV (mean of 30%) was acceptable when the estimated composite CV for the group of individuals was 120% (62). They suggested multiple blood sampling to establish an individual’s baseline CRP. In one study, three measurements at monthly intervals were recommended to define an individual’s steady-state concentration, provided there is no intercurrent infection (57). However, recently it has been shown that two independent measurements of CRP or total cholesterol, 3 months apart, enabled classification of up to 90% of individuals into the exact or immediately adjacent quartile (55). Additional analyses of these data have shown that >95% of individuals would be classified in the exact tertile of risk or vary by one tertile based on the newly recommended cut-points (Fig. 4 ). Continued skepticism surrounding the issue of intraindividual variation remains (63)(64), but the CDC/AHA expert panel concluded that the mean of two independent measurements (fasting or nonfasting) of CRP, taken at least 2 weeks apart, should be used to establish a person’s risk of future coronary events. When CRP is >10 mg/L, CRP measurement should be repeated in 2 weeks to avoid misclassification because of an asymptomatic inflammatory response or a subclinical infection (65).



View larger version (14K):
[in this window]
[in a new window]
  		Figure 4. Within-person variability: comparison of hs-CRP () with total cholesterol (Tchol; {blacksquare}).

Adapted with permission from Okene et al. (55).

Lifestyle (exercise, smoking, obesity, alcohol, antiinflammatory drugs, hormone therapy).
There are limited studies investigating the effect of exercise on CRP concentrations. Strenuous exercise has been shown to increase CRP concentrations. In a group of 30 male marathon runners, median CRP concentrations increased from a pre-race concentration of 1.1 mg/L to a post-race concentration of 4.0 mg/L and increased further (to 22.7 mg/L) 24 h after the race ended (66). An inverse association between CRP concentration and degree of cardiorespiratory fitness was observed in a group of 722 men, with the highest adjusted CRP concentrations in the lowest fitness quintile (67). Among 3638 apparently healthy adults over the age of 40 years, a higher frequency of physical activity was associated with significantly lower odds of having increased CRP or white blood cell concentrations after adjustment for several potential confounding factors (68). The authors of the study concluded that the antiinflammatory effects of routine exercise might help to attenuate CHD risk.

Numerous studies have documented significant correlation between CRP and smoking (28)(69)(70)(71)(72)(73)(74)(75). In general, CRP concentrations increase among smokers with increased cigarette consumption (72). In the elderly, CRP concentrations are associated with lifetime exposure to cigarette smoke. This association is independent of cessation, suggesting that some of the smoking-related damage may be irreversible (73). These data support the hypothesis that CRP is primarily related to lifetime exposure (pack-years) and not to years since cessation of smoking. Furthermore, CRP concentrations were reported to have doubled in current smokers compared with never-smokers, and although they decreased with time since smoking cessation, CRP concentrations remained increased more than 10 years after smoking cessation compared with values for never-smokers (74). In the Physicians’ Health Study (14) and the Women’s Health Study (69), CRP was a good predictor of future MI in both smokers and nonsmokers. In the Helsinki Heart Study, smokers in the highest CRP quartile had a relative risk for CHD of 8.6 compared with 1.6 for smokers in the lowest CRP quartile (75).

Higher CRP concentrations have been found among individuals with increased body mass index (76)(77). The relationships between CRP concentrations and measures of obesity have been reported to be consistent with in vivo release of IL-6 from adipose tissue (78). In fact, nearly one-fourth of IL-6 produced in vivo originates from adipose tissue (79) and is thought to modify adipocyte glucose, lipid metabolism, and body weight (80)(81). In children 10–11 years of age, adiposity was the major determinant of CRP concentrations, with values nearly threefold higher in the top fifth of the ponderal index than in the bottom fifth (82). Moreover, among healthy 2- to 3-year-old children enrolled in a dietary study, obesity was associated with higher fasting insulin concentrations, which in turn were associated with increased CRP concentrations (83). Several studies have clearly demonstrated that significant weight reduction is associated with decreased concentrations of CRP, several cytokines, and adhesion molecules, thus indicating a reduction in the entire inflammatory state of an individual (84)(85).

Among patients with a first MI, alcohol use was associated with CRP concentrations, with never-users having higher concentrations compared with regular drinkers. However, no difference in concentrations was found among control individuals (86). In 17 patients who drank for more than 3 weeks, the median CRP concentrations decreased from 6 to 4 mg/L one week after alcohol withdrawal (70). Recently, Albert et al. (87) showed that moderate alcohol consumption was associated with lower CRP concentrations compared with no or occasional alcohol intake, suggesting that alcohol may attenuate CHD risk in part through antiinflammatory mechanisms. Furthermore, data from prospective studies have shown that IL-6 and tumor necrosis factor-{alpha} receptors 1 and 2 are lower in moderate drinkers than nondrinkers, further suggesting antiinflammatory effects of alcohol (88).

Although it has been shown that aspirin significantly reduces (~60%) the incidence of MI in men with increased CRP concentration, its effect on CRP concentration is uncertain. The authors of one report observed no change in CRP concentrations measured in healthy volunteers receiving aspirin for 7 days (325 or 81 mg/day) (71), whereas the authors of a separate investigation noted a significant decrease in CRP concentration among patients with stable angina pectoris receiving aspirin for 21 days (300 mg/day) (28). The authors of another study reported no appreciable change in CRP concentrations in 57 healthy adults who received 81 mg/day, 81 mg every 3 days, or 325 mg every 3 days for 1 month; however, a sharp reduction in thromboxane ß2 was noted (89). Clearly, larger studies with longer duration of aspirin use are needed to fully determine the effect of this drug on CRP concentration. It is unknown whether periodic or chronic antiinflammatory drug use should be discontinued before blood collection, and if so, how long individuals should be off their medication beforehand. Additional research into this subject would seem appropriate.

Both lovastatin and pravastatin have been shown to reduce coronary events in individuals with increased CRP concentration (30)(90), suggesting an antiinflammatory effect of this class of drugs. Laboratory studies have further confirmed the antiinflammatory effect of statin drugs (91)(92). Several studies have shown that pravastatin, lovastatin, atorvastatin, simvastatin, and cerivastatin lower CRP concentrations by ~15–20% and that the decrease does not appear to be dose related or correlated with LDL-cholesterol (14)(29)(30)(93)(94). It is thought that the CRP reduction may be mediated by reduced monocyte expression of IL-6 and tumor necrosis factor-{alpha} (92).

Both randomized clinical trials and cross-sectional studies have shown that hormone replacement therapy increases serum CRP concentrations by two- to threefold (95)(96)(97). In the Women’s Health Study, those who received hormone replacement therapy had median CRP concentrations twice as high as those who did not receive therapy and age-matched males (96). In an investigation of postmenopausal women enrolled in a trial to evaluate the effects of oral conjugated estrogen and droloxifene, estrogen treatment produced significantly higher IL-6 and CRP concentrations but a slight decrease in soluble E-selectin. In contrast, droloxifene had no effect on CRP and IL-6, but did produce a significant decrease in the concentration of E-selectin. The clinical implications of the mixed profile of both pro- and antiinflammatory effects remain to be elucidated and underscore the need for continuing investigation of selective estrogen replacement modulators (98). The authors of a recent prospective study observed no effect of exogenous androgen therapy on serum inflammatory markers (including CRP) and concluded that a gender difference may exist regarding the effects of estrogen on serum inflammatory markers (99).

Other conditions.
The mean serum CRP concentrations in 15 healthy adult males increased by about 65% after changing altitude from sea level to over 3600 m (100). In evaluating the effect of high altitude on blood chemistries, it is important to adjust for the significant increase in hemoglobin concentration as a result of height. The authors of a study of pregnant women found slightly higher CRP concentrations in pregnant than in nonpregnant women, but they did not observe changes associated with gestational age (101).

Specimen collection
Fasting.
Very little data exist comparing specimens drawn fasting and postprandially for hs-CRP. However, there are well-documented studies showing changes in lipids after a fatty meal. Therefore, in assays that depend on optical clarity, such as turbidimetry and nephelometry, fasting before sampling may be needed.

Time of collection.
It is important to establish whether CRP exhibits a circadian rhythm to determine whether an optimal time for sample collection is necessary for the purpose of assessing future coronary risk. Interest in the diurnal variation of CRP is further stimulated by the fact that proinflammatory cytokines such as IL-6, which stimulates CRP synthesis, exhibit diurnal variation (102)(103). A recent study has shown no evidence of diurnal variation for CRP from hourly blood samples collected from 13 healthy adults (104). The relatively long half-life of CRP (19 h) may have blunted the circadian effect of IL-6.

Specimen type.
Most immunoassays are suitable for work with either serum or plasma; however, data comparing these two fluids are a commonly unrecognized source of variability in CRP assays. It has been reported that the use of EDTA- or citrated plasma specimens produced differences of -12% and -16% in hs-CRP concentration compared with serum (105); the osmotic shifting effect of the anticoagulant on erythrocytes was listed as the likely explanation for the observed discrepancy. Moreover, in an evaluation of a small bench-top device using heparin-plasma and EDTA-whole-blood samples compared directly with serum samples, Deming regression analysis for the comparison between serum and heparin plasma yielded a slope of 0.99, whereas the comparison between serum and EDTA whole blood gave a slope of 0.90 (106). These investigators also noted that twice as many results were one quartile lower with whole-blood specimens compared with serum. In contrast, no significant differences were found when serum, heparin-, and EDTA-plasma samples were simultaneously collected from a single stick in 25 patients (107). In light of the conflicting information, the CDC/AHA committee has stated that there is a need for additional comparisons of hs-CRP assays between serum and plasma samples collected in heparin or in EDTA (42).

Time and temperature of storage.
CRP has been shown to be stable at 4 °C for 60 days (108). No significant effect of storage in liquid nitrogen for 6 months was seen in CRP concentrations on samples collected from apparently healthy individuals (46). In a study of long-term storage, no significant changes in CRP concentrations were seen in healthy individuals or individuals with an APR when serum or plasma was stored at -70 °C for more than 20 years (109). Ideally, when samples are to be stored long term, they should be dispensed into cryotubes with minimal air space and stored at -20 °C or colder in a noncycling freezer. On removal, samples should be thawed slowly in either a freezer or refrigerator, depending on the initial storage condition, and mixed by gentle inversion before use.

analytic variability
Laboratory methodology.
Various commercial methods have been developed to measure CRP in serum and plasma. In the past, many laboratories have used the semiquantitative latex agglutination assay for estimation of the extent of inflammation. However, the lack of assay sensitivity and subjective interpretation make correlation with clinical disease activity difficult. Throughout the 1970s and 1980s, more reliable methods, including nephelometry and turbidimetry, began to appear (110). The advantage of these assays is that they are fully automated, rapid, and reproducible; however, most have a lower detection limit around 5 mg/L, which precludes their use in risk assessment for CHD, where significant changes in the range of 0.5–3.5 mg/L have been reported (14)(15).

Investigators have worked to improve the performance of CRP assays through the use of fluorescent, luminescent, or radioactive adducts to antibodies to enhance the immunoprecipitate and ameliorate the signal. Most of these approaches provide for the reliable measurement of CRP in apparently healthy individuals, but they tend to be more laborious and expensive to perform. An alternative design has been to amplify the light-scattering properties of the antigen-antibody complex by covalently coupling latex particles to a specific antibody. This approach has been very successful and has achieved widespread appeal among clinical laboratories because of the flexibility of chemistry analyzers for turbidimetric applications.

With the increased availability of hs-CRP immunoassays, much discussion concerning their performance and clinical utility has arisen (107)(111)(112)(113). Indeed, different studies using various assays have shown significant discrepancy in reported results and emphasized the need for additional standardization (114). Furthermore, as different clinical applications for CRP have evolved, some laboratories may be required to use two different assays depending on the clinical concern, certainly a potential source of confusion for both the clinician and the laboratorian. Luckily, the third generation of latex-based CRP assays are able to measure CRP over a very wide range of concentrations (~0.1–200 mg/L) (115). The use of different size latex particles has made such measurements feasible.

Detection limit.
As indicated earlier, individuals would be classified into categories of risk on the basis of tertiles of CRP. Therefore, for clinical utility, the assay of interest must be able to reliably measure hs-CRP at least at the lowest cut-point (1 mg/L). Assays used for population-based studies and clinical research, however, should be able to measure hs-CRP concentrations at much lower concentrations, such as 0.15 mg/L (2.5th percentile of the reference population). It has recently been shown that of the nine second-generation hs-CRP assays examined, all had a functional sensitivity <=0.3 mg/L and five had a lower limit of quantification <=0.2 mg/L (112). For clinical utility, all second- and third-generation methods for hs-CRP appear to have the desired sensitivity.

Precision.
The extent to which replicate analyses of a sample agree with each other, usually expressed as imprecision (CV), is the result of combined variables, including antibody affinity, specimen dilution (i.e., degree of turbidity), instrument performance (i.e., lamp deterioration), and operator technique. Determining the precision usually involves multiple measurements of a given specimen at a discrete concentration. A potential problem with this approach is that the imprecision between two or three discrete intervals requires interpolation. An alternative approach is to construct an imprecision profile (116) using specimens of various concentrations (Fig. 5 ). This approach provides valuable information on the assay’s working range, and when sufficient points are included, an estimate of the assay’s functional sensitivity can be determined (117). It was suggested that for hs-CRP, the within-laboratory total imprecision should be <10% across the linear range of the assay (113). In a recent evaluation, only five of the nine hs-CRP methods examined met this criterion (Table 4 ) (112), which underscores the need for more precise assays.



View larger version (17K):
[in this window]
[in a new window]
  		Figure 5. Imprecision profile for a hypothetical hs-CRP assay.

In this example, the dashed horizontal line reflects a CV limit of 10%. The intersection of this line with the fitted data reflects the lower and upper concentration limits for the assay (1.6 and 27.3 mg/L, respectively). The assay’s functional sensitivity (defined as the CRP concentration corresponding to a 20% CV) is 0.8 mg/L and would preclude the use of this assay for estimating CHD risk.


View this table:
[in this window]
[in a new window]
  		Table 4. Reproducibility (CV, %) of hs-CRP methods.1

Antigen excess.
In light-scattering immunoassays, as the antigen concentration increases beyond the equivalence point, smaller immune complexes are formed, which leads to a diminished signal. This diminished signal may correspond to an antigen concentration in antibody excess or beyond the equivalence point in antigen excess. This problem is common with analytes such as CRP, for which there is a very wide pathologic range of concentrations. Among the nine hs-CRP assays evaluated, antigen excess was detected in three of them (112). Manufacturers of hs-CRP tests should work to eliminate these effects; meanwhile, laboratorians should exercise care in selecting and evaluating assays for prozone effects.

Matrix effects.
Matrix effects typically lead to consistent differences between results obtained with different matrices, such as that between serum and the matrix used to prepare a calibrator. These differences might also include variation in optical clarity, protein structure (e.g., monomeric vs the native pentameric protein), and binding to other proteins. In the presence of differences in the protein itself, variations in antibody specificity and reactivity could also contribute to divergence in assayed values. Among nine different hs-CRP methods traceable to the internationally certified reference material (CRM 470), differences ranging from -31% to +28% were seen for a single serum specimen with a concentration of 0.5 mg/L (112). The authors concluded that matrix effects among the various calibrators were a likely factor contributing to the lack of agreement among the methods.

Curve-fitting algorithms.
Most hs-CRP methods use some sort of curve-fitting routine to determine concentrations in patient sera. Before implementing a hs-CRP assay, the laboratory should validate the curve-fitting algorithm for goodness of fit to ensure they do not introduce imprecision or bias into the reported value. Validation techniques involving the back-calculation of calibrator concentrations are relatively straightforward to perform and can reveal concentration-dependent differences. In general, multipoint calibration methods produce more accurate and precise results than single or two-point calibration curves (118). An evaluation of four hs-CRP assays revealed a "bend" in the regression line at ~2.5 mg/L for the Roche, Technoclone, and Biokit methods compared with the Dade Behring assay using the Behring Nephelometer II (111). The bend was attributed to the opposite nonlinearity of the immunoturbidimetric assays compared with the Dade Behring assay. By deriving separate regression equations for concentrations above and below 2.5 mg/L, the authors achieved harmonization of patient results among the different assays. However, the deficiencies identified by these studies stress the need for careful evaluation of all hs-CRP assays, especially at the clinical cut-points.

Method correlation studies.
Method correlation studies are routinely performed to assess the agreement between two methods. Most published studies include the slope and intercept from simple least-squares regression as well as the correlation coefficient (r) to gauge how well the methods agree. Unfortunately, this approach can be misleading because it assumes that the comparison method is without error (119). An alternative strategy is to use Deming regression analysis, which minimizes the error in both the comparison and test methods. In addition, care must be exercised in interpreting a high r value because this does not necessarily indicate good agreement in the range of clinical importance. A preferable approach to interpreting correlation data is to use the Bland–Altman method (120) and an evaluation of the percentage rather than absolute differences between two methods because the SD is frequently concentration-dependent (121). The example presented in Fig. 6 illustrates the advantages of this approach quite clearly. The CDC has embarked on an ambitious effort to develop a reference method for CRP based on liquid chromatography–tandem mass spectrometry that will serve as the accuracy base for future assay evaluations.



View larger version (15K):
[in this window]
[in a new window]
  		Figure 6. Correlation of the of the Behring hs-CRP assay on the Behring Nephelometer Analyzer (BNA) with the Beckman Coulter CRPH assay on the Immage Immunochemistry system.

Adapted with permission from Davis et al. (132). (Top), scatter plot of the two assay systems with Deming regression analysis. (Bottom), Bland–Altman difference plot of the same data reveals significant differences throughout the assay range.

Reference materials.
The vast majority of hs-CRP immunoassays are calibrated to either the WHO 1st International Reference Preparation for C-reactive Protein Immunoassay (85/506), introduced in 1986 (122), or CRM 470, introduced in 1993(123). The value assigned to CRM 470 was derived from WHO IRP 85/506 by use of a very high-precision transfer protocol (51). Despite the availability of valid reference materials, there have been several reports of bias-related problems attributed to standardization or to poor value transfer by the manufacturer (43)(112)(113). In addition, lack of commutability among different assays (e.g., nephelometric vs turbidimetric) from a single manufacturer can arise when the manufacturer’s calibrators and controls are not compared directly with CRM 470 in each assay system (124).

Standardization of hs-CRP assays.
As indicated above, several reports that examined the performance of hs-CRP methods indicated a discrepancy among reported results and suggested the need for further standardization. Agreement among the various hs-CRP methodologies is essential considering that individual patient results will be interpreted within the context of nationally established cut-points. To address this issue, the CDC has initiated a standardization program in which manufacturers of all hs-CRP reagents worldwide have been invited to participate. Phase I of this project aims to identify a suitable reference material. One suggested reason for the lack of agreement in CRP concentrations among the different manufacturers is the relatively high concentrations of CRP in the two primary reference materials (39.3 mg/L for CRM 470 and 49 mg/L for WHO 85/506) because they were developed for use in more traditional applications (125). However, a recent study showed that when an initial 1-in-4 dilution of CRM 470 was made in the manufacturer’s diluent, commutability with both the manufacturer’s calibrator and with dilutions of serum pools could be achieved (126). The study also demonstrated that a previously described value-transfer protocol endorsed by the IFCC could be used with success (51)(127). The main advantages of CRM 470 include the fact that it is universally used by the diagnostic industry for serum protein calibration, it is very stable, it is free from interferents (rheumatoid factor, lipids, and monoclonal proteins), and it is available in large quantities (128). Furthermore, a common reference material that meets the needs of traditional and high-sensitivity applications has obvious practical benefits from the perspective of diagnostic manufacturers. Recent work by the CDC standardization committee on hs-CRP confirmed that CRM 470 performed comparably to the other candidate materials and that it should be used in phase II, which will seek to harmonize various hs-CRP assays in conjunction with a standard value-transfer protocol (129).

Quality-assessment schemes.
The majority of external quality-assessment schemes in use reflect traditional (nonenhanced) CRP assays. Most published reports reflect unusually large within- and among-manufacturer CVs that are concentration-dependent. Data from the 2000–2001 College of American Pathologists (CAP) proficiency program revealed that at CRP concentrations of 28 and 56 mg/L, the mean among-manufacturer CV was 12% compared with 34% at 2.9 mg/L. These data are consistent with a recent study that showed that at CRP concentrations <6 mg/L, among-manufacturer CVs ranged from 30% to 60% and within-manufacturer (among laboratory) CVs were as high as 160% (130). At CRP concentrations >20 mg/L, the among-manufacturer CV was <20% and the within-manufacturer CVs were 3–15%. Similar findings have been reported in earlier studies from Belgium (117) and the United Kingdom (130). These results not only underscore the need for more sensitive immunoassays, but virtually preclude the use of most traditional assays for estimating CHD risk (124).

In 2002, the CAP introduced a proficiency survey program for hs-CRP methods. Among the different method peer groups, CVs were 21–90% at a CRP concentration of 0.24 mg/L; 6–23% at 1.46 mg/L, and 5–11% at 11.3 mg/L (Fig. 7A ). Accuracy (calculated as the peer group mean to the all-method mean) ranged from 45% to 225% at 0.24 mg/L; from 91% to 116% at 1.46 mg/L, and from 96% to 114% at 11.3 mg/L (Fig. 7B ). The large CVs and observed differences among results clearly support the need for improved standardization efforts.



View larger version (20K):
[in this window]
[in a new window]
  		Figure 7. Data derived from the 2002 CAP Cardiac Risk Survey program results for seven different methods.

(A), precision results for survey specimen 1 ({square}; overall mean hs-CRP = 0.24 mg/L; overall CV = 120%); survey specimen 2 (; overall mean hs-CRP = 1.46 mg/L, overall CV = 16%); and survey specimen 3 ({blacksquare}; overall mean hs-CRP = 11.3 mg/L, overall CV = 11%). (B), accuracy plot for the same specimens expressed as the ratio of the peer-group mean to the all-method mean.


   Conclusions
Top
 Abstract
 Introduction
 Structure
 Genetics
 Physiologic Function
 Clinical Significance
 Sources of Variability in...
 Conclusions
References
 
Although CRP was discovered more than 70 years ago, its clinical utility has been hampered by lack of understanding of its function and by the difficulties associated with accurate quantification. With growing awareness of the role of CRP in health and disease, we will undoubtedly see a continued expansion in the use of this test. The utility of CRP to predict future coronary events in apparently healthy individuals and to assess prognosis in patients with acute coronary syndromes has renewed interest in its measurement. Clinical guidelines for the utility of CRP in the primary prevention of CHD as well as in patients with stable coronary disease or acute coronary syndrome have now become available. Better control of preanalytic and analytic sources of variations will undoubtedly lead to improvement in CRP measurements. hs-CRP methods are currently available, and many appear to be reliable. However, additional standardization of these methods is needed to improve accuracy and assure harmonization among reported CRP results.


   Footnotes
 
1 Nonstandard abbreviations: APR, acute-phase response; CRP, C-reactive protein; IL, interleukin; MI, myocardial infarction; CHD, coronary heart disease; AHA, American Heart Association; hs, high sensitivity; CRM, Certified Reference Material; and CAP, College of American Pathologists.


   References
Top
 Abstract
 Introduction
 Structure
 Genetics
 Physiologic Function
 Clinical Significance
 Sources of Variability in...
 Conclusions
 References
 

  1. Tillet WS, Francis T. Serological reaction in pneumonia with a non-protein somatic fraction of pneumococcus. J Exp Med 1930;52:561-571.[Abstract]
  2. MacLeod CM, Avery OT. The occurrence during acute infections of a protein not normally present in the blood. II. Isolation and properties of the reactive protein. J Exp Med 1943;73:183-191.
  3. Lofstrom G. Comparison between the reaction of acute phase serum with pneumococcus C-polysaccharide and with pneumococcus type 27. Br J Exp Pathol 1944;25:21-26.
  4. Kushner I. C-reactive protein in rheumatology. Arthritis Rheum 1991;34:1065-1068.[ISI][Medline] [Order article via Infotrieve]
  5. Pepys MB. C-reactive protein fifty years on. Lancet 1981;1:653-657.[ISI][Medline] [Order article via Infotrieve]
  6. Kilpatrick JM, Volanakis JE. Molecular genetics, structure, and function of C-reactive protein. Immunol Res 1991;10:43-53.[ISI][Medline] [Order article via Infotrieve]
  7. Zee RY, Ridker PM. Polymorphism in the human C-reactive protein (CRP) gene, plasma concentrations of CRP, and the risk of future arterial thrombosis. Atherosclerosis 2002;162:217-219.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  8. Pankow JS, Folsom AR, Cushman M, Borecki IB, Hopkins PN, Eckfeldt JH, et al. Familial and genetic determinants of systemic markers of inflammation: the NHLBI family heart study. Atherosclerosis 2001;154:681-689.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  9. Castell JV, Gomez-Lechon MJ, David M, Fabra R, Trullenque R, Heinrich PC. Acute-phase response of human hepatocytes: regulation of acute-phase protein synthesis by interleukin-6. Hepatology 1990;12:1179-1186.[ISI][Medline] [Order article via Infotrieve]
  10. Johnson AM, Rohlfs E, Silverman LM. Proteins. Burtis CA Ashwood ER eds. Tietz textbook of clinical chemistry, 3rd ed 1999:477-540 WB Saunders Philadelphia. .
  11. Bienvenu J, Whicher JT, Aguzzi F. C-reactive protein. Ritchie RF Navolotskaia O eds. Serum proteins in clinical medicine 1996:7.01.01-7.01.06 Maine Printing Group Portland, ME. .
  12. Hoffmann JA, Kafatos FC, Janeway CA, Ezekowitz RA. Phylogenetic perspectives in innate immunity. Science 1999;284:1313-1318.[Abstract/Free Full Text]
  13. Clyne B, Olshaker JS. The C-reactive protein. J Emerg Med 1999;6:1019-1025.
  14. Ridker PM, Cushman M, Stampfer MJ, Tracy RP, Hennekens CH. Inflammation, aspirin, and the risk of cardiovascular disease in apparently healthy men. N Engl J Med 1997;336:973-979.[Abstract/Free Full Text]
  15. Ridker PM, Hennekens CH, Buring JE, Rifai N. C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N Engl J Med 2000;342:836-843.[Abstract/Free Full Text]
  16. Köenig W, Sund M, Fröhlich M, Fischer HG, Lowel H, Doring A, et al. C-Reactive protein, a sensitive marker of inflammation, predicts future risk of coronary heart disease in initially healthy middle-aged men: results from the MONICA (Monitoring Trends and Determinants in Cardiovascular Disease) Augsburg Cohort Study, 1984 to 1992. Circulation 1999;99:237-242.[Abstract/Free Full Text]
  17. Rost NS, Wolf PA, Kase CS, Kelly-Hayes M, Silbershatz H, Massaro JM, et al. Plasma concentration of C-reactive protein and risk of ischemic stroke and transient ischemic attack: the Framingham study. Stroke 2001;32:2575-2579.[Abstract/Free Full Text]
  18. Albert CM, Ma J, Rifai N, Stampfer MJ, Ridker PM. Prospective study of C-reactive protein, homocysteine, and plasma lipid levels as predictors of sudden cardiac death. Circulation 2002;105:2595-2599.[Abstract/Free Full Text]
  19. Ridker PM, Stampfer MJ, Rifai N. Novel risk factors for systemic atherosclerosis: a comparison of C-reactive protein, fibrinogen, homocysteine, lipoprotein(a), and standard cholesterol screening as predictors of peripheral arterial disease. JAMA 2001;285:2481-2485.[Abstract/Free Full Text]
  20. Ridker PM, Hennekens CH, Roitman-Johnson B, Stampfer MJ, Allen J. Plasma concentration of soluble intercellular adhesion molecule 1 and risks of future myocardial infarction in apparently healthy men. Lancet 1998;351:88-92.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  21. Ridker PM, Buring JE, Rifai N. Soluble P-selectin and the risk of future cardiovascular events. Circulation 2001;103:491-495.[Abstract/Free Full Text]
  22. Kuller LH, Tracy RP, Shaten J, Meilahn EN. Relation of C-reactive protein and coronary heart disease in the MRFIT nested case-control study. Multiple Risk Factor Intervention Trial. Am J Epidemiol 1996;144:537-547.[Abstract/Free Full Text]
  23. Danesh J, Whincup P, Walker M, Lennon L, Thomson A, Appleby P, et al. Low grade inflammation and coronary heart disease: prospective study and updated meta-analyses. BMJ 2000;321:199-204.[Abstract/Free Full Text]
  24. Ridker PM, Glynn RJ, Hennekens CH. C-reactive protein adds to the predictive value of total and HDL cholesterol in determining risk of first myocardial infarction. Circulation 1998;97:2007-2011.[Abstract/Free Full Text]
  25. Ridker PM, Rifai N, Rose L, Buring JE, Cook NR. Comparison of C-reactive protein and low-density lipoprotein cholesterol levels in the prediction of first cardiovascular events. N Engl J Med 2002;347:1557-1565.[Abstract/Free Full Text]
  26. Ridker PM, Buring JE, Cook NR, Rifai N. C-reactive protein, the metabolic syndrome, and risk of incident cardiovascular events: an 8-year follow-up of 14 719 initially healthy American women. Circulation 2003;107:391-397.[Abstract/Free Full Text]
  27. Kennon S, Price CP, Mills PG, Ranjadayalan K, Cooper J, Clarke H, et al. The effect of aspirin on C-reactive protein as a marker of risk in unstable angina. J Am Coll Cardiol 2001;37:1266-1270.[Abstract/Free Full Text]
  28. Ikonomidis I, Andreotti F, Economou E, Stefanadis C, Toutouzas P, Nihoyannopoulos P. Increased proinflammatory cytokines in patients with chronic stable angina and their reduction by aspirin. Circulation 1999;100:793-798.[Abstract/Free Full Text]
  29. Ridker PM, Rifai N, Clearfield M, Downs JR, Weis SE, Miles JS, et al. Measurement of C-reactive protein for the targeting of statin therapy in the primary prevention of acute coronary events. N Engl J Med 2001;344:1959-1965.[Abstract/Free Full Text]
  30. Ridker PM, Rifai N, Pfeffer MA, Sacks FM, Moye LA, Goldman S, et al. Inflammation, pravastatin, and the risk of coronary events after myocardial infarction in patients with average cholesterol levels. Cholesterol and Recurrent Events (CARE) Investigators. Circulation 1998;98:839-844.[Abstract/Free Full Text]
  31. Morrow DA, Rifai N, Antman EM, Weiner DL, McCabe CH, Cannon CP, et al. C-reactive protein is a potent predictor of mortality independently of and in combination with troponin T in acute coronary syndromes: a TIMI 11A substudy. Thrombolysis in Myocardial Infarction. J Am Coll Cardiol 1998;31:1460-1465.[Abstract/Free Full Text]
  32. Lagrand WK, Visser CA, Hermens WT, Niessen HW, Verheugt FW, Wolbink GJ, et al. C-reactive protein as a cardiovascular risk factor: more than an epiphenomenon?. Circulation 1999;100:96-102.[Abstract/Free Full Text]
  33. Reynolds GD, Vance RP. C-reactive protein immunohistochemical localization in normal and atherosclerotic human aortas. Arch Pathol Lab Med 1987;111:265-269.[ISI][Medline] [Order article via Infotrieve]
  34. Liuzzo G, Baisucci LM, Gallimore JR, Caligiuri G, Buffon A, Rebuzzi AG, et al. Enhanced inflammatory response in patients with preinfarction unstable angina. J Am Coll Cardiol 1999;34:1696-1703.[Abstract/Free Full Text]
  35. Liuzzo G, Biasucci LM, Gallimore JR, Grillo RL, Rebuzzi AG, Pepys MB, et al. The prognostic value of C-reactive protein and serum amyloid A protein in severe unstable angina. N Engl J Med 1994;331:417-424.[Abstract/Free Full Text]
  36. Biasucci LM, Vitelli A, Liuzzo G, Altamura S, Caligiuri G, Monaco C, et al. Elevated levels of interleukin-6 in unstable angina. Circulation 1996;94:874-877.[Abstract/Free Full Text]
  37. Pasceri V, Willerson JT, Yeh ET. Direct proinflammatory effect of C-reactive protein on human endothelial cells. Circulation 2000;102:2165-2168.[Abstract/Free Full Text]
  38. Pasceri V, Cheng JS, Willerson JT, Yeh ET, Chang J. Modulation of C-reactive protein-mediated monocyte chemoattractant protein-1 induction in human endothelial cells by anti-atherosclerosis drugs. Circulation 2001;103:2531-2534.[Abstract/Free Full Text]
  39. Verma S, Li SH, Badiwala MV, Weisel RD, Fedak PW, Li RK, et al. Endothelin antagonism and interleukin-6 inhibition attenuate the proatherogenic effects of C-reactive protein. Circulation 2002;105:1890-1896.[Abstract/Free Full Text]
  40. Venugopal SK, Devaraj S, Yuhanna I, Shaul P, Jialal I. Demonstration that C-reactive protein decreases eNOS expression and bioactivity in human aortic endothelial cells. Circulation 2002;106:1439-1441.[Abstract/Free Full Text]
  41. Devaraj S, Xu DY, Jialal I. C-reactive protein increases plasminogen activator inhibitor-1 expression and activity in human aortic endothelial cells: implications for the metabolic syndrome and atherothrombosis. Circulation 2003;107:398-404.[Abstract/Free Full Text]
  42. Pearson TA, Mensah GA, Alexander RW, Anderson JL, Cannon RO, III, Criqui M, et al. Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation 2003;107:499-511.[Free Full Text]
  43. Ledue TB, Weiner DL, Sipe JD, Poulin SE, Collins MF, Rifai N. Analytical evaluation of particle-enhanced immunonephelometric assays for C-reactive protein, serum amyloid A and mannose-binding protein in human serum. Ann Clin Biochem 1998;35:745-753.
  44. Erlandsen EJ, Randers E. Reference interval for serum C-reactive protein in healthy blood donors using the Dade Behring N Latex CRP mono assay. Scand J Clin Lab Invest 2000;60:37-43.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  45. Ford ES, Giles WH, Myers GL, Mannino DM. Population distribution of high-sensitivity C-reactive protein among US men: findings from National Health and Nutrition Examination Survey 1999–2000. Clin Chem 2003;49:686-690.[Free Full Text]
  46. Rifai N, Ridker PM. Population distributions of C-reactive protein in apparently healthy men and women in the United States: implication for clinical interpretation [Technical Brief]. Clin Chem 2003;49:666-669.[Free Full Text]
  47. Imhof A, Frohlich M, Loewel H, Helbecque N, Woodward M, Amouyel P, et al. Distributions of C-reactive protein measured by high-sensitivity assays in apparently healthy men and women from different populations in Europe [Technical Brief]. Clin Chem 2003;49:669-672.[Free Full Text]
  48. Yamada S, Gotoh T, Nakashima Y, Kayaba K, Ishikawa S, Nago N, et al. Distribution of serum C-reactive protein and its association with atherosclerotic risk factors in a Japanese population: Jichi Medical School Cohort Study. Am J Epidemiol 2001;153:1183-1190.[Abstract/Free Full Text]
  49. Chambers JC, Eda S, Bassett P, Karim Y, Thompson SG, Gallimore JR, et al. C-reactive protein, insulin resistance, central obesity, and coronary heart disease risk in Indian Asians from the United Kingdom compared with European whites. Circulation 2001;104:145-150.[Abstract/Free Full Text]
  50. Wener MH, Daum PR, McQuillan GM. The influence of age, sex, and race on the upper reference limit of serum C-reactive protein concentration. J Rheumatol 2000;27:2351-2359.[ISI][Medline] [Order article via Infotrieve]
  51. Baudner S, Bienvenu J, Blirup-Jensen S, Carlstrom A, Johnson AM, Ward AM, et al. The certification of a matrix reference material for immunochemical measurement of 14 human serum proteins. CRM 470 1993:1-172 Community Bureau of Reference, Commission of the European Communities Brussels. .
  52. Hutchinson WL, Köenig W, Frohlich M, Sund M, Lowe GD, Pepys MB. Immunoradiometric assay of circulating C-reactive protein: age-related values in the adult general population. Clin Chem 2000;46:934-938.[Abstract/Free Full Text]
  53. Herberth B, Siest G, Henny J. High-sensitivity C-reactive protein (CRP) reference intervals in the elderly. Clin Chem Lab Med 2001;11:1169-1170.
  54. Crawford VL, Sweeney O, Coyle PV, Halliday IM, Stout RW. The relationship between elevated fibrinogen and markers of infection: a comparison of seasonal cycles. QJM 2000;93:745-750.[Abstract/Free Full Text]
  55. Ockene IS, Matthews CE, Rifai N, Ridker PM, Reed G, Stanek E. Variability and classification accuracy of serial high-sensitivity C-reactive protein measurements in healthy adults. Clin Chem 2001;47:444-450.[Abstract/Free Full Text]
  56. Clark GH, Fraser CG. Biological variation of acute phase proteins. Ann Clin Biochem 1993;30:373-376.
  57. Macy EM, Hayes TE, Tracy RP. Variability in the measurement of C-reactive protein in health