"Data are from refs. 18 and 20.
"Data are from refs. 18 and 20.
studies have suggested that at least small differences in the distribution of CRP by ethnicity do exist (21). However, at present gender- or ethnic-specific cut points for CRP are not advocated.
Seasonal and temporal. There are limited data on CRP concentrations and seasonal cycles. Although one small study reported a slight variability in CRP concentration between
-2-10 1 2 Agreement Between First and Second Measures Fig. 4. Within-person variability for CRP vs total cholesterol (TC). (With permission from ref. 26.)
winter and summer (22), the SEASON study, which was specifically designed to examine seasonal changes in cardiovascular biomarkers, showed no consistent pattern of change in CRP concentrations (23).
It is well established that CRP has a relatively large biological variability. Therefore, the impact of this variability on the clinical utility of CRP in CHD risk assessment has been the subject of great interest. In two studies, the within-subject coefficient of variation (CV) ranged from 42 to 63% and the between-subject CV ranged from 76 to 92% (24,25), leading investigators to question the reliability ofthis marker in heart disease risk assessment. However, another study showed that the large intraindividual CV (averaging 30%) was acceptable when the estimated composite CV for the group of individuals was 120% (26). Nevertheless, given this biological variability, multiple blood sampling is recommended to establish an individual's baseline CRP. Some have advocated three independent determinations (25); however, findings from the SEASON trial showed that two independent measurements of CRP or total cholesterol, 3 months apart, enabled classification of up to 90% of subjects into the exact or immediately adjacent quartile (23). Additional analyses of these data have shown that >95% of subjects would be classified in the exact tertile of risk or vary by one tertile using the newly recommended cut points (27) (Fig. 4). Furthermore, in the Cholesterol and Recurrent Events trial, the age-adjusted correlation between two CRP measurements from blood samples drawn 5 yr apart was 0.6—again, a value comparable with that of cholesterol and other lipid parameters (28). Although continued skepticism surrounding the issue ofintraindividual variation remains (29), the AHA/ CDC expert panel concluded that the average oftwo independent measurements (fasting or nonfasting) of CRP taken at least 2 wk apart should be used to establish an individual's risk of CHD (9). Although it was initially recommended to repeat the measurement when the CRP concentration exceeds 10 mg/L (23), subsequent evidence suggests that the association of CRP with risk extends well beyond that range of concentration (30).
Exercise, body mass, and other lifestyle behavior. Several physiological conditions and lifestyle behaviors, such as exercise, obesity, cigarette smoking, and alcohol consumption, are known to affect CRP concentrations. Strenuous exercise, such as running a marathon, increases CRP concentrations acutely; however, an inverse association between CRP
concentration and levels of cardiorespiratory fitness has also been reported (31). In addition, a higher frequency of physical activity is associated with significantly lower odds of having increased CRP (32).
A positive association between CRP concentrations and body mass index has been clearly demonstrated (33,34). 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. 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 (35,36).
Numerous studies have documented an increased CRP concentration with cigarette smoking (37,38). This association is independent of cessation, suggesting that some of the smoking-related damage may be irreversible. In both the Physicians' Health Study (39) and the Women's Health Study (38), CRP was a good predictor of future myocardial infarction (MI) in both smokers and nonsmokers.
Moderate alcohol consumption is associated with lower CRP concentrations compared with no or occasional alcohol intake, suggesting that alcohol may attenuate the risk of CHD in part through anti-inflammatory mechanisms. Furthermore, IL-6 and TNF-a receptors 1 and 2 are lower in moderate drinkers than nondrinkers, further implicating anti-inflammatory effects of alcohol (40).
Medications. Several pharmacological agents and treatment modalities, including HRT, aspirin, and statins, influence CRP concentrations. In randomized clinical trials and cross-sectional studies, HRT increased serum CRP concentrations by two- to threefold (41-43). One study showed that although estrogen treatment resulted in significantly higher CRP concentrations, droloxifene exhibited no effect (44). In a prospective study in men, no effect of exogenous androgen therapy on serum inflammatory markers (including CRP) was observed, leading to the conclusion that a gender difference may exist regarding the effects of estrogen on serum inflammatory markers (45).
Although the effect of aspirin on reducing the incidence of MI in men with increased CRP (~60%) is clear (39), its effect on CRP concentration is uncertain. One study found a significant reduction in CRP concentration with the use of aspirin (46), but several others reported no change (47-49). By contrast, the statins have uniformly been shown to reduce concentrations of CRP by 15-20% (28,50,51).
There is no evidence to suggest that samples must be collected in the fasting state for the measurement of CRP (Table 3). However, certain assays are affected by optical clarity, and fasting before sampling may be needed in the presence of severe hypertriglycerid-emia. Furthermore, CRP does not appear to exhibit a circadian rhythm and, therefore, there is no need to standardize the time for sample collection to assess the risk of CHD (52). Concern regarding diurnal variation of CRP stems from the fact that IL-6, which stimulates CRP synthesis, exhibits significant diurnal variation (53). However, the lack of diurnal variation of CRP may be owing to its relatively long half-life (19 h), which probably has blunted the circadian effect of IL-6.
Most commercially available immunoassays are suitable for either serum or plasma. However, the difference between these two fluids with respect to CRP testing is unclear. As expected, because of the osmotic shifting effect of the anticoagulant on erythrocytes, the use of EDTA or citrated plasma specimens resulted in significant biases (>10%) in CRP
Practical Considerations in CRP Testing
• What type of specimen should be used? Serum values equivalent to heparinized plasma EDTA <5% and citrate <10% than serum
• Should the patient be fasting?
No: fasting (6.6 ± 13.5 mg/L) vs nonfasting (6.3 ± 14 mg/L) (n = 10; p < 0.37)
• Should the time for sample collection be standardized? No: no circadian rhythm exhibited for CRP
• How should CRP results be reported and interpreted?
In mg/L: values <1 = low, 1-3 = average, >3 = high risk of CHD
concentration when compared with serum in one study (54). By contrast, no difference in CRP values was seen using heparinized and serum samples (55). Another study, however, showed no significant differences when serum and heparin- or EDTA-plasma samples were simultaneously collected from a single stick in 25 patients (56). Additional studies are needed to clarify this important issue.
CRP has been shown to be stable at 4°C for 60 d (57), and no significant changes in its concentration were seen in samples stored at -70°C for more than 20 yr (58) or in liquid nitrogen for up to 6 mo (17).
Was this article helpful?